Publications by authors named "Pekka Ruusuvuori"

41 Publications

Building a central repository landmarks a new era for artificial intelligence-assisted digital pathology development in Europe.

Eur J Cancer 2021 Apr 20;150:31-32. Epub 2021 Apr 20.

Institute of Biomedicine, University of Turku, Turku, Finland. Electronic address:

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ejca.2021.03.018DOI Listing
April 2021

Convolutional Neural Network-Based Artificial Intelligence for Classification of Protein Localization Patterns.

Biomolecules 2021 02 11;11(2). Epub 2021 Feb 11.

Faculty of Medicine and Health Technology, Tampere University, FI-33520 Tampere, Finland.

Identifying localization of proteins and their specific subpopulations associated with certain cellular compartments is crucial for understanding protein function and interactions with other macromolecules. Fluorescence microscopy is a powerful method to assess protein localizations, with increasing demand of automated high throughput analysis methods to supplement the technical advancements in high throughput imaging. Here, we study the applicability of deep neural network-based artificial intelligence in classification of protein localization in 13 cellular subcompartments. We use deep learning-based on convolutional neural network and fully convolutional network with similar architectures for the classification task, aiming at achieving accurate classification, but importantly, also comparison of the networks. Our results show that both types of convolutional neural networks perform well in protein localization classification tasks for major cellular organelles. Yet, in this study, the fully convolutional network outperforms the convolutional neural network in classification of images with multiple simultaneous protein localizations. We find that the fully convolutional network, using output visualizing the identified localizations, is a very useful tool for systematic protein localization assessment.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/biom11020264DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7916854PMC
February 2021

Interobserver reproducibility of perineural invasion of prostatic adenocarcinoma in needle biopsies.

Virchows Arch 2021 Feb 3. Epub 2021 Feb 3.

Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland.

Numerous studies have shown a correlation between perineural invasion (PNI) in prostate biopsies and outcome. The reporting of PNI varies widely in the literature. While the interobserver variability of prostate cancer grading has been studied extensively, less is known regarding the reproducibility of PNI. A total of 212 biopsy cores from a population-based screening trial were included in this study (106 with and 106 without PNI according to the original pathology reports). The glass slides were scanned and circulated among four pathologists with a special interest in urological pathology for assessment of PNI. Discordant cases were stained by immunohistochemistry for S-100 protein. PNI was diagnosed by all four observers in 34.0% of cases, while 41.5% were considered to be negative for PNI. In 24.5% of cases, there was a disagreement between the observers. The kappa for interobserver variability was 0.67-0.75 (mean 0.73). The observations from one participant were compared with data from the original reports, and a kappa for intraobserver variability of 0.87 was achieved. Based on immunohistochemical findings among discordant cases, 88.6% had PNI while 11.4% did not. The most common diagnostic pitfall was the presence of bundles of stroma or smooth muscle. It was noted in a few cases that collagenous micronodules could be mistaken for a nerve. The distance between cancer and nerve was another cause of disagreement. Although the results suggest that the reproducibility of PNI may be greater than that of prostate cancer grading, there is still a need for improvement and standardization.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s00428-021-03039-zDOI Listing
February 2021

Single cell characterization of B-lymphoid differentiation and leukemic cell states during chemotherapy in ETV6-RUNX1-positive pediatric leukemia identifies drug-targetable transcription factor activities.

Genome Med 2020 Nov 20;12(1):99. Epub 2020 Nov 20.

Institute of Biomedicine, School of Medicine, University of Eastern Finland, Yliopistonranta 1, FI-70211, Kuopio, Finland.

Background: Tight regulatory loops orchestrate commitment to B cell fate within bone marrow. Genetic lesions in this gene regulatory network underlie the emergence of the most common childhood cancer, acute lymphoblastic leukemia (ALL). The initial genetic hits, including the common translocation that fuses ETV6 and RUNX1 genes, lead to arrested cell differentiation. Here, we aimed to characterize transcription factor activities along the B-lineage differentiation trajectory as a reference to characterize the aberrant cell states present in leukemic bone marrow, and to identify those transcription factors that maintain cancer-specific cell states for more precise therapeutic intervention.

Methods: We compared normal B-lineage differentiation and in vivo leukemic cell states using single cell RNA-sequencing (scRNA-seq) and several complementary genomics profiles. Based on statistical tools for scRNA-seq, we benchmarked a workflow to resolve transcription factor activities and gene expression distribution changes in healthy bone marrow lymphoid cell states. We compared these to ALL bone marrow at diagnosis and in vivo during chemotherapy, focusing on leukemias carrying the ETV6-RUNX1 fusion.

Results: We show that lymphoid cell transcription factor activities uncovered from bone marrow scRNA-seq have high correspondence with independent ATAC- and ChIP-seq data. Using this comprehensive reference for regulatory factors coordinating B-lineage differentiation, our analysis of ETV6-RUNX1-positive ALL cases revealed elevated activity of multiple ETS-transcription factors in leukemic cells states, including the leukemia genome-wide association study hit ELK3. The accompanying gene expression changes associated with natural killer cell inactivation and depletion in the leukemic immune microenvironment. Moreover, our results suggest that the abundance of G1 cell cycle state at diagnosis and lack of differentiation-associated regulatory network changes during induction chemotherapy represent features of chemoresistance. To target the leukemic regulatory program and thereby overcome treatment resistance, we show that inhibition of ETS-transcription factors reduced cell viability and resolved pathways contributing to this using scRNA-seq.

Conclusions: Our data provide a detailed picture of the transcription factor activities characterizing both normal B-lineage differentiation and those acquired in leukemic bone marrow and provide a rational basis for new treatment strategies targeting the immune microenvironment and the active regulatory network in leukemia.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s13073-020-00799-2DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7679990PMC
November 2020

Generalized fixation invariant nuclei detection through domain adaptation based deep learning.

IEEE J Biomed Health Inform 2020 Nov 19;PP. Epub 2020 Nov 19.

Nucleus detection is a fundamental task in histological image analysis and an important tool for many follow up analyses. It is known that sample preparation and scanning procedure of histological slides introduce a great amount of variability to the histological images and poses challenges for automated nucleus detection. Here, we studied the effect of histopathological sample fixation on the accuracy of a deep learning based nuclei detection model trained with hematoxylin and eosin stained images. We experimented with training data that includes three methods of fixation; PAXgene, formalin and frozen, and studied the detection accuracy results of various convolutional neural networks. Our results indicate that the variability introduced during sample preparation affects the generalization of a model and should be considered when building accurate and robust nuclei detection algorithms. Our dataset includes over 67 000 annotated nuclei locations from 16 patients and three different sample fixation types. The dataset provides excellent basis for building an accurate and robust nuclei detection model, and combined with unsupervised domain adaptation, the workflow allows generalization to images from unseen domains, including different tissues and images from different labs.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/JBHI.2020.3039414DOI Listing
November 2020

ANHIR: Automatic Non-Rigid Histological Image Registration Challenge.

IEEE Trans Med Imaging 2020 10 7;39(10):3042-3052. Epub 2020 Apr 7.

Automatic Non-rigid Histological Image Registration (ANHIR) challenge was organized to compare the performance of image registration algorithms on several kinds of microscopy histology images in a fair and independent manner. We have assembled 8 datasets, containing 355 images with 18 different stains, resulting in 481 image pairs to be registered. Registration accuracy was evaluated using manually placed landmarks. In total, 256 teams registered for the challenge, 10 submitted the results, and 6 participated in the workshop. Here, we present the results of 7 well-performing methods from the challenge together with 6 well-known existing methods. The best methods used coarse but robust initial alignment, followed by non-rigid registration, used multiresolution, and were carefully tuned for the data at hand. They outperformed off-the-shelf methods, mostly by being more robust. The best methods could successfully register over 98% of all landmarks and their mean landmark registration accuracy (TRE) was 0.44% of the image diagonal. The challenge remains open to submissions and all images are available for download.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2020.2986331DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7584382PMC
October 2020

Glioblastoma Multiforme Stem Cell Cycle Arrest by Alkylaminophenol Through the Modulation of EGFR and CSC Signaling Pathways.

Cells 2020 03 10;9(3). Epub 2020 Mar 10.

Molecular Signaling Lab, Faculty of Medicine and Health Technology, Tampere University, P.O. Box 553, 33101 Tampere, Finland.

Cancer stem cells (CSCs), a small subpopulation of cells existing in the tumor microenvironment promoting cell proliferation and growth. Targeting the stemness of the CSC population would offer a vital therapeutic opportunity. 3,4-Dihydroquinolin-1(2)-yl)(-tolyl)methyl)phenol (THTMP), a small synthetic phenol compound, is proposed to play a significant role in controlling the CSC proliferation and survival. We assessed the potential therapeutic effects of THTMP on glioblastoma multiforme (GBM) and its underlying mechanism in various signaling pathways. To fully comprehend the effect of THTMP on the CSCs, CD133 GBM stem cell (GSC) and CD133 GBM Non-stem cancer cells (NSCC) population from LN229 and SNB19 cell lines was used. Cell cycle arrest, apoptosis assay and transcriptome analysis were performed for individual cell population. THTMP strongly inhibited NSCC and in a subtle way for GSC in a time-dependent manner and inhibit the resistance variants better than that of temozolomide (TMZ). THTMP arrest the CSC cell population at both G1/S and G2/M phase and induce ROS-mediated apoptosis. Gene expression profiling characterize THTMP as an inhibitor of the p53 signaling pathway causing DNA damage and cell cycle arrest in CSC population. We show that the THTMP majorly affects the EGFR and CSC signaling pathways. Specifically, modulation of key genes involved in Wnt, Notch and Hedgehog, revealed the significant role of THTMP in disrupting the CSCs' stemness and functions. Moreover, THTMP inhibited cell growth, proliferation and metastasis of multiple mesenchymal patient-tissue derived GBM-cell lines. THTMP arrests GBM stem cell cycle through the modulation of EGFR and CSC signaling pathways.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/cells9030681DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7140667PMC
March 2020

3D-Printed Whole Prostate Models with Tumor Hotspots Using Dual-Extruder Printer.

Annu Int Conf IEEE Eng Med Biol Soc 2019 Jul;2019:2867-2871

3D printing has emerged as a popular technology in various biomedical applications. Physical models of anatomical structures concretize the digital representations and can be used for teaching and analysis. In this study we combine 3D histology with 3D printing, creating realistic physical models of tissues with hotspots of interest. As an example we use mouse prostates containing tumors. Surface meshes are created from binary masks of HE-stained serial sections of mouse prostates and manually annotated tumor areas. Sections are interpolated to expand sparse image stacks for smoother results. Fiji, Meshlab and Tinkercad are used for mesh creation and processing. Objects are printed with Prusa-based dual-extruder printer enabling different colors for tumors and the surrounding prostate tissue. Our 3D-printed mouse prostates appear realistic and tumors located at the edges of the organ are clearly visible. When transparent filament is used, the tumor hotspots are visible even when they are inside the prostate.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC.2019.8857068DOI Listing
July 2019

Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study.

Lancet Oncol 2020 02 8;21(2):222-232. Epub 2020 Jan 8.

Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.

Background: An increasing volume of prostate biopsies and a worldwide shortage of urological pathologists puts a strain on pathology departments. Additionally, the high intra-observer and inter-observer variability in grading can result in overtreatment and undertreatment of prostate cancer. To alleviate these problems, we aimed to develop an artificial intelligence (AI) system with clinically acceptable accuracy for prostate cancer detection, localisation, and Gleason grading.

Methods: We digitised 6682 slides from needle core biopsies from 976 randomly selected participants aged 50-69 in the Swedish prospective and population-based STHLM3 diagnostic study done between May 28, 2012, and Dec 30, 2014 (ISRCTN84445406), and another 271 from 93 men from outside the study. The resulting images were used to train deep neural networks for assessment of prostate biopsies. The networks were evaluated by predicting the presence, extent, and Gleason grade of malignant tissue for an independent test dataset comprising 1631 biopsies from 246 men from STHLM3 and an external validation dataset of 330 biopsies from 73 men. We also evaluated grading performance on 87 biopsies individually graded by 23 experienced urological pathologists from the International Society of Urological Pathology. We assessed discriminatory performance by receiver operating characteristics and tumour extent predictions by correlating predicted cancer length against measurements by the reporting pathologist. We quantified the concordance between grades assigned by the AI system and the expert urological pathologists using Cohen's kappa.

Findings: The AI achieved an area under the receiver operating characteristics curve of 0·997 (95% CI 0·994-0·999) for distinguishing between benign (n=910) and malignant (n=721) biopsy cores on the independent test dataset and 0·986 (0·972-0·996) on the external validation dataset (benign n=108, malignant n=222). The correlation between cancer length predicted by the AI and assigned by the reporting pathologist was 0·96 (95% CI 0·95-0·97) for the independent test dataset and 0·87 (0·84-0·90) for the external validation dataset. For assigning Gleason grades, the AI achieved a mean pairwise kappa of 0·62, which was within the range of the corresponding values for the expert pathologists (0·60-0·73).

Interpretation: An AI system can be trained to detect and grade cancer in prostate needle biopsy samples at a ranking comparable to that of international experts in prostate pathology. Clinical application could reduce pathology workload by reducing the assessment of benign biopsies and by automating the task of measuring cancer length in positive biopsy cores. An AI system with expert-level grading performance might contribute a second opinion, aid in standardising grading, and provide pathology expertise in parts of the world where it does not exist.

Funding: Swedish Research Council, Swedish Cancer Society, Swedish eScience Research Center, EIT Health.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/S1470-2045(19)30738-7DOI Listing
February 2020

Phosphorylation of NFATC1 at PIM1 target sites is essential for its ability to promote prostate cancer cell migration and invasion.

Cell Commun Signal 2019 11 15;17(1):148. Epub 2019 Nov 15.

Department of Biology, University of Turku, Vesilinnantie 5, FI-20500, Turku, Finland.

Background: Progression of prostate cancer from benign local tumors to metastatic carcinomas is a multistep process. Here we have investigated the signaling pathways that support migration and invasion of prostate cancer cells, focusing on the role of the NFATC1 transcription factor and its post-translational modifications. We have previously identified NFATC1 as a substrate for the PIM1 kinase and shown that PIM1-dependent phosphorylation increases NFATC1 activity without affecting its subcellular localization. Both PIM kinases and NFATC1 have been reported to promote cancer cell migration, invasion and angiogenesis, but it has remained unclear whether the effects of NFATC1 are phosphorylation-dependent and which downstream targets are involved.

Methods: We used mass spectrometry to identify PIM1 phosphorylation target sites in NFATC1, and analysed their functional roles in three prostate cancer cell lines by comparing phosphodeficient mutants to wild-type NFATC1. We used luciferase assays to determine effects of phosphorylation on NFAT-dependent transcriptional activity, and migration and invasion assays to evaluate effects on cell motility. We also performed a microarray analysis to identify novel PIM1/NFATC1 targets, and validated one of them with both cellular expression analyses and in silico in clinical prostate cancer data sets.

Results: Here we have identified ten PIM1 target sites in NFATC1 and found that prevention of their phosphorylation significantly decreases the transcriptional activity as well as the pro-migratory and pro-invasive effects of NFATC1 in prostate cancer cells. We observed that also PIM2 and PIM3 can phosphorylate NFATC1, and identified several novel putative PIM1/NFATC1 target genes. These include the ITGA5 integrin, which is differentially expressed in the presence of wild-type versus phosphorylation-deficient NFATC1, and which is coexpressed with PIM1 and NFATC1 in clinical prostate cancer specimens.

Conclusions: Based on our data, phosphorylation of PIM1 target sites stimulates NFATC1 activity and enhances its ability to promote prostate cancer cell migration and invasion. Therefore, inhibition of the interplay between PIM kinases and NFATC1 may have therapeutic implications for patients with metastatic forms of cancer.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12964-019-0463-yDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6858710PMC
November 2019

Cytokeratin-Supervised Deep Learning for Automatic Recognition of Epithelial Cells in Breast Cancers Stained for ER, PR, and Ki-67.

IEEE Trans Med Imaging 2020 02 7;39(2):534-542. Epub 2019 Aug 7.

Immunohistochemistry (IHC) of ER, PR, and Ki-67 are routinely used assays in breast cancer diagnostics. Determination of the proportion of stained cells (labeling index) should be restricted on malignant epithelial cells, carefully avoiding tumor infiltrating stroma and inflammatory cells. Here, we developed a deep learning based digital mask for automated epithelial cell detection using fluoro-chromogenic cytokeratin-Ki-67 double staining and sequential hematoxylin-IHC staining as training material. A partially pre-trained deep convolutional neural network was fine-tuned using image batches from 152 patient samples of invasive breast tumors. Validity of the trained digital epithelial cell masks was studied with 366 images captured from 98 unseen samples, by comparing the epithelial cell masks to cytokeratin images and by visual evaluation of the brightfield images performed by two pathologists. A good discrimination of epithelial cells was achieved (AUC of mean ROC = 0.93; defined as the area under mean receiver operating characteristics), and well in concordance with pathologists' visual assessment (4.01/5 and 4.67/5). The effect of epithelial cell masking on the Ki-67 labeling index was substantial. 52 tumor images initially classified as low proliferation (Ki-67 < 14%) without epithelial cell masking were re-classified as high proliferation (Ki-67 ≥ 14%) after applying the deep learning based epithelial cell mask. The digital epithelial cell masks were found applicable also to IHC of ER and PR. We conclude that deep learning can be applied to detect carcinoma cells in breast cancer samples stained with conventional brightfield IHC.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2019.2933656DOI Listing
February 2020

Iterative unsupervised domain adaptation for generalized cell detection from brightfield z-stacks.

BMC Bioinformatics 2019 Feb 15;20(1):80. Epub 2019 Feb 15.

Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland.

Background: Cell counting from cell cultures is required in multiple biological and biomedical research applications. Especially, accurate brightfield-based cell counting methods are needed for cell growth analysis. With deep learning, cells can be detected with high accuracy, but manually annotated training data is required. We propose a method for cell detection that requires annotated training data for one cell line only, and generalizes to other, unseen cell lines.

Results: Training a deep learning model with one cell line only can provide accurate detections for similar unseen cell lines (domains). However, if the new domain is very dissimilar from training domain, high precision but lower recall is achieved. Generalization capabilities of the model can be improved with training data transformations, but only to a certain degree. To further improve the detection accuracy of unseen domains, we propose iterative unsupervised domain adaptation method. Predictions of unseen cell lines with high precision enable automatic generation of training data, which is used to train the model together with parts of the previously used annotated training data. We used U-Net-based model, and three consecutive focal planes from brightfield image z-stacks. We trained the model initially with PC-3 cell line, and used LNCaP, BT-474 and 22Rv1 cell lines as target domains for domain adaptation. Highest improvement in accuracy was achieved for 22Rv1 cells. F-score after supervised training was only 0.65, but after unsupervised domain adaptation we achieved a score of 0.84. Mean accuracy for target domains was 0.87, with mean improvement of 16 percent.

Conclusions: With our method for generalized cell detection, we can train a model that accurately detects different cell lines from brightfield images. A new cell line can be introduced to the model without a single manual annotation, and after iterative domain adaptation the model is ready to detect these cells with high accuracy.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12859-019-2605-zDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6376647PMC
February 2019

Comparative analysis of tissue reconstruction algorithms for 3D histology.

Bioinformatics 2018 09;34(17):3013-3021

Faculty of Medicine and Life Sciences, University of Tampere, Tampere, Finland.

Motivation: Digital pathology enables new approaches that expand beyond storage, visualization or analysis of histological samples in digital format. One novel opportunity is 3D histology, where a three-dimensional reconstruction of the sample is formed computationally based on serial tissue sections. This allows examining tissue architecture in 3D, for example, for diagnostic purposes. Importantly, 3D histology enables joint mapping of cellular morphology with spatially resolved omics data in the true 3D context of the tissue at microscopic resolution. Several algorithms have been proposed for the reconstruction task, but a quantitative comparison of their accuracy is lacking.

Results: We developed a benchmarking framework to evaluate the accuracy of several free and commercial 3D reconstruction methods using two whole slide image datasets. The results provide a solid basis for further development and application of 3D histology algorithms and indicate that methods capable of compensating for local tissue deformation are superior to simpler approaches.

Availability And Implementation: Code: https://github.com/BioimageInformaticsTampere/RegBenchmark. Whole slide image datasets: http://urn.fi/urn: nbn: fi: csc-kata20170705131652639702.

Supplementary Information: Supplementary data are available at Bioinformatics online.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1093/bioinformatics/bty210DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6129300PMC
September 2018

Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women With Breast Cancer.

JAMA 2017 12;318(22):2199-2210

Munich Business School, Munich, Germany.

Importance: Application of deep learning algorithms to whole-slide pathology images can potentially improve diagnostic accuracy and efficiency.

Objective: Assess the performance of automated deep learning algorithms at detecting metastases in hematoxylin and eosin-stained tissue sections of lymph nodes of women with breast cancer and compare it with pathologists' diagnoses in a diagnostic setting.

Design, Setting, And Participants: Researcher challenge competition (CAMELYON16) to develop automated solutions for detecting lymph node metastases (November 2015-November 2016). A training data set of whole-slide images from 2 centers in the Netherlands with (n = 110) and without (n = 160) nodal metastases verified by immunohistochemical staining were provided to challenge participants to build algorithms. Algorithm performance was evaluated in an independent test set of 129 whole-slide images (49 with and 80 without metastases). The same test set of corresponding glass slides was also evaluated by a panel of 11 pathologists with time constraint (WTC) from the Netherlands to ascertain likelihood of nodal metastases for each slide in a flexible 2-hour session, simulating routine pathology workflow, and by 1 pathologist without time constraint (WOTC).

Exposures: Deep learning algorithms submitted as part of a challenge competition or pathologist interpretation.

Main Outcomes And Measures: The presence of specific metastatic foci and the absence vs presence of lymph node metastasis in a slide or image using receiver operating characteristic curve analysis. The 11 pathologists participating in the simulation exercise rated their diagnostic confidence as definitely normal, probably normal, equivocal, probably tumor, or definitely tumor.

Results: The area under the receiver operating characteristic curve (AUC) for the algorithms ranged from 0.556 to 0.994. The top-performing algorithm achieved a lesion-level, true-positive fraction comparable with that of the pathologist WOTC (72.4% [95% CI, 64.3%-80.4%]) at a mean of 0.0125 false-positives per normal whole-slide image. For the whole-slide image classification task, the best algorithm (AUC, 0.994 [95% CI, 0.983-0.999]) performed significantly better than the pathologists WTC in a diagnostic simulation (mean AUC, 0.810 [range, 0.738-0.884]; P < .001). The top 5 algorithms had a mean AUC that was comparable with the pathologist interpreting the slides in the absence of time constraints (mean AUC, 0.960 [range, 0.923-0.994] for the top 5 algorithms vs 0.966 [95% CI, 0.927-0.998] for the pathologist WOTC).

Conclusions And Relevance: In the setting of a challenge competition, some deep learning algorithms achieved better diagnostic performance than a panel of 11 pathologists participating in a simulation exercise designed to mimic routine pathology workflow; algorithm performance was comparable with an expert pathologist interpreting whole-slide images without time constraints. Whether this approach has clinical utility will require evaluation in a clinical setting.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1001/jama.2017.14585DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5820737PMC
December 2017

Feasibility of Prostate PAXgene Fixation for Molecular Research and Diagnostic Surgical Pathology: Comparison of Matched Fresh Frozen, FFPE, and PFPE Tissues.

Am J Surg Pathol 2018 Jan;42(1):103-115

Prostate Cancer Research Center, Faculty of Medicine and Life Sciences and BioMediTech Institute.

Advances in prostate cancer biology and diagnostics are dependent upon high-fidelity integration of clinical, histomorphologic, and molecular phenotypic findings. In this study, we compared fresh frozen, formalin-fixed paraffin-embedded (FFPE), and PAXgene-fixed paraffin-embedded (PFPE) tissue preparation methods in radical prostatectomy prostate tissue from 36 patients and performed a preliminary test of feasibility of using PFPE tissue in routine prostate surgical pathology diagnostic assessment. In addition to comparing histology, immunohistochemistry, and general measures of DNA and RNA integrity in each fixation method, we performed functional tests of DNA and RNA quality, including targeted Miseq RNA and DNA sequencing, and implemented methods to relate DNA and RNA yield and quality to quantified DNA and RNA picogram nuclear content in each tissue volume studied. Our results suggest that it is feasible to use PFPE tissue for routine robot-assisted laparoscopic prostatectomy surgical pathology diagnostics and immunohistochemistry, with the benefit of significantly improvedDNA and RNA quality and RNA picogram yield per nucleus as compared with FFPE tissue. For fresh frozen, FFPE, and PFPE tissues, respectively, the average Genomic Quality Numbers were 7.9, 3.2, and 6.2, average RNA Quality Numbers were 8.7, 2.6, and 6.3, average DNA picogram yields per nucleus were 0.41, 0.69, and 0.78, and average RNA picogram yields per nucleus were 1.40, 0.94, and 2.24. These findings suggest that where DNA and/or RNA analysis of tissue is required, and when tissue size is small, PFPE may provide important advantages over FFPE. The results also suggest several interesting nuances including potential avenues to improve RNA quality in FFPE tissues and confirm recent suggestions that some DNA sequence artifacts associated with FFPE can be avoided.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1097/PAS.0000000000000961DOI Listing
January 2018

In Vivo Expression of miR-32 Induces Proliferation in Prostate Epithelium.

Am J Pathol 2017 Nov 19;187(11):2546-2557. Epub 2017 Aug 19.

Prostate Cancer Research Center, Faculty of Medicine and Life Sciences and BioMediTech, University of Tampere, Tampere, Finland; Fimlab Laboratories, Tampere University Hospital, Tampere, Finland. Electronic address:

miRNAs are important regulators of gene expression and are often deregulated in cancer. We have previously shown that miR-32 is an androgen receptor-regulated miRNA overexpressed in castration-resistant prostate cancer and that miR-32 can improve prostate cancer cell growth in vitro. To assess the effects of miR-32 in vivo, we developed transgenic mice overexpressing miR-32 in the prostate. The study indicated that transgenic miR-32 expression increases replicative activity in the prostate epithelium. We further observed an aging-associated increase in the incidence of goblet cell metaplasia in the prostate epithelium. Furthermore, aged miR-32 transgenic mice exhibited metaplasia-associated prostatic intraepithelial neoplasia at a low frequency. When crossbred with mice lacking the other allele of tumor-suppressor Pten (miR-32xPten mice), miR-32 expression increased both the incidence and the replicative activity of prostatic intraepithelial neoplasia lesions in the dorsal prostate. The miR-32xPten mice also demonstrated increased goblet cell metaplasia compared with Pten mice. By performing a microarray analysis of mouse prostate tissue to screen downstream targets and effectors of miR-32, we identified RAC2 as a potential, and clinically relevant, target of miR-32. We also demonstrate down-regulation of several interesting, potentially prostate cancer-relevant genes (Spink1, Spink5, and Casp1) by miR-32 in the prostate tissue. The results demonstrate that miR-32 increases proliferation and promotes metaplastic transformation in mouse prostate epithelium, which may promote neoplastic alterations in the prostate.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ajpath.2017.07.012DOI Listing
November 2017

Metastasis detection from whole slide images using local features and random forests.

Cytometry A 2017 06 20;91(6):555-565. Epub 2017 Apr 20.

BioMediTech and Faculty of Medicine and Life Sciences, University of Tampere, Tampere, Finland.

Digital pathology has led to a demand for automated detection of regions of interest, such as cancerous tissue, from scanned whole slide images. With accurate methods using image analysis and machine learning, significant speed-up, and savings in costs through increased throughput in histological assessment could be achieved. This article describes a machine learning approach for detection of cancerous tissue from scanned whole slide images. Our method is based on feature engineering and supervised learning with a random forest model. The features extracted from the whole slide images include several local descriptors related to image texture, spatial structure, and distribution of nuclei. The method was evaluated in breast cancer metastasis detection from lymph node samples. Our results show that the method detects metastatic areas with high accuracy (AUC = 0.97-0.98 for tumor detection within whole image area, AUC = 0.84-0.91 for tumor vs. normal tissue detection) and that the method generalizes well for images from more than one laboratory. Further, the method outputs an interpretable classification model, enabling the linking of individual features to differences between tissue types. © 2017 International Society for Advancement of Cytometry.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/cyto.a.23089DOI Listing
June 2017

Strong FGFR3 staining is a marker for FGFR3 fusions in diffuse gliomas.

Neuro Oncol 2017 Sep;19(9):1206-1216

BioMediTech Institute and Faculty of Medicine and Life Sciences, University of Tampere, Tampere, Finland; Department of Signal Processing, Tampere University of Technology, Tampere, Finland; Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, Texas; Science Center, Tampere University Hospital, Tampere, Finland; Fimlab Laboratories Ltd., Tampere University Hospital, Tampere, Finland; Unit of Neurosurgery, Tampere University Hospital, Tampere, Finland; Pori unit, Tampere University of Technology, Pori, Finland; Department of Pathology, University of Tampere, Tampere, Finland; Department of Cancer Biology, Comprehensive Cancer Center of Wake Forest Baptist Medical Center, Winston-Salem, North Carolina.

Background: Inhibitors of fibroblast growth factor receptors (FGFRs) have recently arisen as a promising treatment option for patients with FGFR alterations. Gene fusions involving FGFR3 and transforming acidic coiled-coil protein 3 (TACC3) have been detected in diffuse gliomas and other malignancies, and fusion-positive cases have responded well to FGFR inhibition. As high FGFR3 expression has been detected in fusion-positive tumors, we sought to determine the clinical significance of FGFR3 protein expression level as well as its potential for indicating FGFR3 fusions.

Methods: We performed FGFR3 immunohistochemistry on tissue microarrays containing 676 grades II-IV astrocytomas and 116 grades II-III oligodendroglial tumor specimens. Fifty-one cases were further analyzed using targeted sequencing.

Results: Moderate to strong FGFR3 staining was detected in gliomas of all grades, was more common in females, and was associated with poor survival in diffuse astrocytomas. Targeted sequencing identified FGFR3-TACC3 fusions and an FGFR3-CAMK2A fusion in 10 of 15 strongly stained cases, whereas no fusions were found in 36 negatively to moderately stained cases. Fusion-positive cases were predominantly female and negative for IDH and EGFR/PDGFRA/MET alterations. These and moderately stained cases show lower MIB-1 proliferation index than negatively to weakly stained cases. Furthermore, stronger FGFR3 expression was commonly observed in malignant tissue regions of lower cellularity in fusion-negative cases. Importantly, subregional negative FGFR3 staining was also observed in a few fusion-positive cases.

Conclusions: Strong FGFR3 protein expression is indicative of FGFR3 fusions and may serve as a clinically applicable predictive marker for treatment regimens based on FGFR inhibitors.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1093/neuonc/nox028DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5570261PMC
September 2017

Analysis of spatial heterogeneity in normal epithelium and preneoplastic alterations in mouse prostate tumor models.

Sci Rep 2017 03 20;7:44831. Epub 2017 Mar 20.

Prostate Cancer Research Center, Faculty of Medicine and Life Sciences and BioMediTech, University of Tampere, Tampere, Finland.

Cancer involves histological changes in tissue, which is of primary importance in pathological diagnosis and research. Automated histological analysis requires ability to computationally separate pathological alterations from normal tissue with all its variables. On the other hand, understanding connections between genetic alterations and histological attributes requires development of enhanced analysis methods suitable also for small sample sizes. Here, we set out to develop computational methods for early detection and distinction of prostate cancer-related pathological alterations. We use analysis of features from HE stained histological images of normal mouse prostate epithelium, distinguishing the descriptors for variability between ventral, lateral, and dorsal lobes. In addition, we use two common prostate cancer models, Hi-Myc and Pten+/- mice, to build a feature-based machine learning model separating the early pathological lesions provoked by these genetic alterations. This work offers a set of computational methods for separation of early neoplastic lesions in the prostates of model mice, and provides proof-of-principle for linking specific tumor genotypes to quantitative histological characteristics. The results obtained show that separation between different spatial locations within the organ, as well as classification between histologies linked to different genetic backgrounds, can be performed with very high specificity and sensitivity.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/srep44831DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5357939PMC
March 2017

Virtual cell imaging: A review on simulation methods employed in image cytometry.

Cytometry A 2016 12 6;89(12):1057-1072. Epub 2016 Dec 6.

Institute of Biosciences and Medical Technology - BioMediTech, University of Tampere, Tampere, Finland.

The simulations of cells and microscope images thereof have been used to facilitate the development, selection, and validation of image analysis algorithms employed in cytometry as well as for modeling and understanding cell structure and dynamics beyond what is visible in the eyepiece. The simulation approaches vary from simple parametric models of specific cell components-especially shapes of cells and cell nuclei-to learning-based synthesis and multi-stage simulation models for complex scenes that simultaneously visualize multiple object types and incorporate various properties of the imaged objects and laws of image formation. This review covers advances in artificial digital cell generation at scales ranging from particles up to tissue synthesis and microscope image simulation methods, provides examples of the use of simulated images for various purposes ranging from subcellular object detection to cell tracking, and discusses how such simulators have been validated. Finally, the future possibilities and limitations of simulation-based validation are considered. © 2016 International Society for Advancement of Cytometry.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/cyto.a.23031DOI Listing
December 2016

Echovirus 1 internalization negatively regulates epidermal growth factor receptor downregulation.

Cell Microbiol 2017 03 20;19(3). Epub 2016 Oct 20.

Department of Biological and Environmental Science/NanoScience Center, University of Jyväskylä, Jyväskylä, Finland.

We have demonstrated previously that the human picornavirus Echovirus 1 (EV1) triggers an infectious internalization pathway that follows closely, but seems to stay separate, from the epidermal growth factor receptor (EGFR) pathway triggered by epidermal growth factor (EGF). Here, we confirmed by using live and confocal microscopy that EGFR and EV1 vesicles are following intimately each other but are distinct entities with different degradation kinetics. We show here that despite being sorted to different pathways and located in distinct endosomes, EV1 inhibits EGFR downregulation. Simultaneous treatment with EV1 and EGF led to an accumulation of EGFR in cytoplasmic endosomes, which was evident already 15 min p.i. and more pronounced after 2 hr p.i. EV1 treatment led to reduced downregulation, which was proven by increased total cellular amount of EGFR. Confocal microscopy studies revealed that EGFR accumulated in large endosomes, presumably macropinosomes, which were not positive for markers of the early, recycling, or late endosomes/lysosomes. Interestingly, EV1 did not have a similar blocking effect on bulk endosomal trafficking or transferrin recycling along the clathrin pathway suggesting that EV1 did not have a general effect on cellular trafficking pathways. Importantly, EGF treatment increased EV1 infection and increased cell viability during infection. Simultaneous EV1 and EGF treatment seemed to moderately enhance phosphorylation of protein kinase C α. Furthermore, similar phenotype of EGFR trafficking could be produced by phorbol 12-myristate 13-acetate treatment, further suggesting that activated protein kinase C α could be contributing to EGFR phenotype. These results altogether demonstrate that EV1 specifically affects EGFR trafficking, leading to EGFR downregulation, which is beneficial to EV1 infection.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1111/cmi.12671DOI Listing
March 2017

Feature-based analysis of mouse prostatic intraepithelial neoplasia in histological tissue sections.

J Pathol Inform 2016 29;7. Epub 2016 Jan 29.

Institute of Biosciences and Medical Technology - BioMediTech, University of Tampere, Tampere, Finland; Fimlab Laboratories, Tampere University Hospital, Tampere, Finland.

This paper describes work presented at the Nordic Symposium on Digital Pathology 2015, in Linköping, Sweden. Prostatic intraepithelial neoplasia (PIN) represents premalignant tissue involving epithelial growth confined in the lumen of prostatic acini. In the attempts to understand oncogenesis in the human prostate, early neoplastic changes can be modeled in the mouse with genetic manipulation of certain tumor suppressor genes or oncogenes. As with many early pathological changes, the PIN lesions in the mouse prostate are macroscopically small, but microscopically spanning areas often larger than single high magnification focus fields in microscopy. This poses a challenge to utilize full potential of the data acquired in histological specimens. We use whole prostates fixed in molecular fixative PAXgene™, embedded in paraffin, sectioned through and stained with H&E. To visualize and analyze the microscopic information spanning whole mouse PIN (mPIN) lesions, we utilize automated whole slide scanning and stacked sections through the tissue. The region of interests is masked, and the masked areas are processed using a cascade of automated image analysis steps. The images are normalized in color space, after which exclusion of secretion areas and feature extraction is performed. Machine learning is utilized to build a model of early PIN lesions for determining the probability for histological changes based on the calculated features. We performed a feature-based analysis to mPIN lesions. First, a quantitative representation of over 100 features was built, including several features representing pathological changes in PIN, especially describing the spatial growth pattern of lesions in the prostate tissue. Furthermore, we built a classification model, which is able to align PIN lesions corresponding to grading by visual inspection to more advanced and mild lesions. The classifier allowed both determining the probability of early histological changes for uncategorized tissue samples and interpretation of the model parameters. Here, we develop quantitative image analysis pipeline to describe morphological changes in histological images. Even subtle changes in mPIN lesion characteristics can be described with feature analysis and machine learning. Constructing and using multidimensional feature data to represent histological changes enables richer analysis and interpretation of early pathological lesions.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.4103/2153-3539.175378DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4763506PMC
March 2016

Recurrent SKIL-activating rearrangements in ETS-negative prostate cancer.

Oncotarget 2015 Mar;6(8):6235-50

Institute of Biosciences and Medical Technology - BioMediTech, University of Tampere, Tampere, Finland.

Prostate cancer is the third most common cause of male cancer death in developed countries, and one of the most comprehensively characterized human cancers. Roughly 60% of prostate cancers harbor gene fusions that juxtapose ETS-family transcription factors with androgen regulated promoters. A second subtype, characterized by SPINK1 overexpression, accounts for 15% of prostate cancers. Here we report the discovery of a new prostate cancer subtype characterized by rearrangements juxtaposing the SMAD inhibitor SKIL with androgen regulated promoters, leading to increased SKIL expression. SKIL fusions were found in 6 of 540 (1.1%) prostate cancers and 1 of 27 (3.7%) cell lines and xenografts. 6 of 7 SKIL-positive cancers were negative for ETS overexpression, suggesting mutual exclusivity with ETS fusions. SKIL knockdown led to growth arrest in PC-3 and LNCaP cell line models of prostate cancer, and its overexpression led to increased invasiveness in RWPE-1 cells. The role of SKIL as a prostate cancer oncogene lends support to recent studies on the role of TGF-β signaling as a rate-limiting step in prostate cancer progression. Our findings highlight SKIL as an oncogene and potential therapeutic target in 1-2% of prostate cancers, amounting to an estimated 10,000 cancer diagnoses per year worldwide.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4467434PMC
http://dx.doi.org/10.18632/oncotarget.3359DOI Listing
March 2015

Flow Cytometry-Based Classification in Cancer Research: A View on Feature Selection.

Cancer Inform 2015 10;14(Suppl 5):75-85. Epub 2016 Apr 10.

Department of Signal Processing, Tampere University of Technology, Tampere, Finland.

In this paper, we study the problem of feature selection in cancer-related machine learning tasks. In particular, we study the accuracy and stability of different feature selection approaches within simplistic machine learning pipelines. Earlier studies have shown that for certain cases, the accuracy of detection can easily reach 100% given enough training data. Here, however, we concentrate on simplifying the classification models with and seek for feature selection approaches that are reliable even with extremely small sample sizes. We show that as much as 50% of features can be discarded without compromising the prediction accuracy. Moreover, we study the model selection problem among the ℓ 1 regularization path of logistic regression classifiers. To this aim, we compare a more traditional cross-validation approach with a recently proposed Bayesian error estimator.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.4137/CIN.S30795DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4827794PMC
April 2016

Unidirectional P-body transport during the yeast cell cycle.

PLoS One 2014 11;9(6):e99428. Epub 2014 Jun 11.

Molecular and Cellular Biology Program, University of Washington, Seattle, Washington, United States of America; Pacific Northwest Diabetes Research Institute, Seattle, Washington, United States of America.

P-bodies belong to a large family of RNA granules that are associated with post-transcriptional gene regulation, conserved from yeast to mammals, and influence biological processes ranging from germ cell development to neuronal plasticity. RNA granules can also transport RNAs to specific locations. Germ granules transport maternal RNAs to the embryo, and neuronal granules transport RNAs long distances to the synaptic dendrites. Here we combine microfluidic-based fluorescent microscopy of single cells and automated image analysis to follow p-body dynamics during cell division in yeast. Our results demonstrate that these highly dynamic granules undergo a unidirectional transport from the mother to the daughter cell during mitosis as well as a constrained "hovering" near the bud site half an hour before the bud is observable. Both behaviors are dependent on the Myo4p/She2p RNA transport machinery. Furthermore, single cell analysis of cell size suggests that PBs play an important role in daughter cell growth under nutrient limiting conditions.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0099428PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4053424PMC
February 2015

Quantitative analysis of dynamic association in live biological fluorescent samples.

PLoS One 2014 11;9(4):e94245. Epub 2014 Apr 11.

Department of Biological and Environmental Science/Nanoscience Center, University of Jyväskylä, Jyväskylä, Finland.

Determining vesicle localization and association in live microscopy may be challenging due to non-simultaneous imaging of rapidly moving objects with two excitation channels. Besides errors due to movement of objects, imaging may also introduce shifting between the image channels, and traditional colocalization methods cannot handle such situations. Our approach to quantifying the association between tagged proteins is to use an object-based method where the exact match of object locations is not assumed. Point-pattern matching provides a measure of correspondence between two point-sets under various changes between the sets. Thus, it can be used for robust quantitative analysis of vesicle association between image channels. Results for a large set of synthetic images shows that the novel association method based on point-pattern matching demonstrates robust capability to detect association of closely located vesicles in live cell-microscopy where traditional colocalization methods fail to produce results. In addition, the method outperforms compared Iterated Closest Points registration method. Results for fixed and live experimental data shows the association method to perform comparably to traditional methods in colocalization studies for fixed cells and to perform favorably in association studies for live cells.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0094245PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3984138PMC
January 2015

Quantitative analysis of colony morphology in yeast.

Biotechniques 2014 Jan;56(1):18-27

Pacific Northwest Diabetes Research Institute, Seattle, WA; Molecular and Cellular Biology Program, University of Washington, Seattle, WA.

Microorganisms often form multicellular structures such as biofilms and structured colonies that can influence the organism's virulence, drug resistance, and adherence to medical devices. Phenotypic classification of these structures has traditionally relied on qualitative scoring systems that limit detailed phenotypic comparisons between strains. Automated imaging and quantitative analysis have the potential to improve the speed and accuracy of experiments designed to study the genetic and molecular networks underlying different morphological traits. For this reason, we have developed a platform that uses automated image analysis and pattern recognition to quantify phenotypic signatures of yeast colonies. Our strategy enables quantitative analysis of individual colonies, measured at a single time point or over a series of time-lapse images, as well as the classification of distinct colony shapes based on image-derived features. Phenotypic changes in colony morphology can be expressed as changes in feature space trajectories over time, thereby enabling the visualization and quantitative analysis of morphological development. To facilitate data exploration, results are plotted dynamically through an interactive Yeast Image Analysis web application (YIMAA; http://yimaa.cs.tut.fi) that integrates the raw and processed images across all time points, allowing exploration of the image-based features and principal components associated with morphological development.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.2144/000114123DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3996921PMC
January 2014

Multi-scale Gaussian representation and outline-learning based cell image segmentation.

BMC Bioinformatics 2013 12;14 Suppl 10:S6. Epub 2013 Aug 12.

Background: High-throughput genome-wide screening to study gene-specific functions, e.g. for drug discovery, demands fast automated image analysis methods to assist in unraveling the full potential of such studies. Image segmentation is typically at the forefront of such analysis as the performance of the subsequent steps, for example, cell classification, cell tracking etc., often relies on the results of segmentation.

Methods: We present a cell cytoplasm segmentation framework which first separates cell cytoplasm from image background using novel approach of image enhancement and coefficient of variation of multi-scale Gaussian scale-space representation. A novel outline-learning based classification method is developed using regularized logistic regression with embedded feature selection which classifies image pixels as outline/non-outline to give cytoplasm outlines. Refinement of the detected outlines to separate cells from each other is performed in a post-processing step where the nuclei segmentation is used as contextual information.

Results And Conclusions: We evaluate the proposed segmentation methodology using two challenging test cases, presenting images with completely different characteristics, with cells of varying size, shape, texture and degrees of overlap. The feature selection and classification framework for outline detection produces very simple sparse models which use only a small subset of the large, generic feature set, that is, only 7 and 5 features for the two cases. Quantitative comparison of the results for the two test cases against state-of-the-art methods show that our methodology outperforms them with an increase of 4-9% in segmentation accuracy with maximum accuracy of 93%. Finally, the results obtained for diverse datasets demonstrate that our framework not only produces accurate segmentation but also generalizes well to different segmentation tasks.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/1471-2105-14-S10-S6DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3750482PMC
March 2014

Leukemia prediction using sparse logistic regression.

PLoS One 2013 30;8(8):e72932. Epub 2013 Aug 30.

Department of Signal Processing, Tampere University of Technology, Tampere, Finland.

We describe a supervised prediction method for diagnosis of acute myeloid leukemia (AML) from patient samples based on flow cytometry measurements. We use a data driven approach with machine learning methods to train a computational model that takes in flow cytometry measurements from a single patient and gives a confidence score of the patient being AML-positive. Our solution is based on an [Formula: see text] regularized logistic regression model that aggregates AML test statistics calculated from individual test tubes with different cell populations and fluorescent markers. The model construction is entirely data driven and no prior biological knowledge is used. The described solution scored a 100% classification accuracy in the DREAM6/FlowCAP2 Molecular Classification of Acute Myeloid Leukaemia Challenge against a golden standard consisting of 20 AML-positive and 160 healthy patients. Here we perform a more extensive validation of the prediction model performance and further improve and simplify our original method showing that statistically equal results can be obtained by using simple average marker intensities as features in the logistic regression model. In addition to the logistic regression based model, we also present other classification models and compare their performance quantitatively. The key benefit in our prediction method compared to other solutions with similar performance is that our model only uses a small fraction of the flow cytometry measurements making our solution highly economical.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0072932PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3758279PMC
April 2014

Bioactive acellular implant induces angiogenesis and adipogenesis and sustained soft tissue restoration in vivo.

Tissue Eng Part A 2012 Dec 17;18(23-24):2568-80. Epub 2012 Aug 17.

Department of Cell Biology, School of Medicine, University of Tampere, Tampere, Finland.

Soft tissue defects resulting from trauma, tumor resection, or congenital causes provide a challenging problem to reconstructive surgery and tissue engineering. Current therapeutic procedures lack the ability to induce rapid formation of neovascularization. Therefore, to date, no adequate application for the reconstruction of soft tissue defects is available. We have previously shown that bioactive factors extracted from adipose tissue (adipose tissue extract [ATE]) induce both adipogenesis and angiogenesis in vitro. These bioactive factors were incorporated into hyaluronan (HA) hydrogel, and the ATE-HA implant-induced angiogenesis and adipogenesis were studied. The developed implant was shown to gradually release the bioactive factors, and the presence of the implant in human adipose stem cell culture was able to induce adipogenic differentiation as evaluated by Oil-red-O staining. In animal experiments, the implants were placed under dorsal subcutis of rodents. Either rat- (rATE, allograft) or human- (hATE, xenograft) derived ATE was incorporated into implants. Local inflammation reactions, angiogenesis, and adipogenesis were followed from 1 week to 40 weeks. Angiogenesis was assessed by microvessel density analysis; adipogenesis was assessed by automated image analysis, and immunological effects by immunostaining and counting inflammatory cells. The key requirements for soft tissue replacement--host compatibility, bioactivity, and sustainability--were all achieved with the novel ATE-HA implant. This acellular implant induced microvessel induction early after implantation and adipose tissue deposition from 12 weeks onward as well as subcutaneous tissue volume increase. The ATE-HA implant was replaced by mature adipose tissue with capillaries, nerve bundles, and healthy connective tissue without local inflammation or capsule formation. The large fat pads remained in tissue until the end of the follow-up time, for 9 months. No adverse effects were detected at the site of implantation, and according to irritating ranking, the ATE-implant was considered to have excellent biocompatibility. The results demonstrate that an acellular HA hydrogel implant induces significant increase in adipogenesis and angiogenesis in vivo compared to the plain HA implant, and ATE has excellent potential for use in tissue engineering for sustained reconstruction of soft tissue defects.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1089/ten.TEA.2011.0724DOI Listing
December 2012