Publications by authors named "Michael Fulham"

135 Publications

Predicting distant metastases in soft-tissue sarcomas from PET-CT scans using constrained hierarchical multi-modality feature learning.

Phys Med Biol 2021 Nov 24. Epub 2021 Nov 24.

The University of Sydney, Sydney, 2006, AUSTRALIA.

Objective: Positron emission tomography-computed tomography (PET-CT) is regarded as the imaging modality of choice for the management of soft-tissue sarcomas (STSs). Distant metastases (DM) are the leading cause of death in STS patients and early detection is important to effectively manage tumors with surgery, radiotherapy and chemotherapy. In this study, we aim to early detect DM in patients with STS using their PET-CT data.

Approach: We derive a new convolutional neural network (CNN) method for early DM detection. The novelty of our method is the introduction of a constrained hierarchical multi-modality feature learning approach to integrate functional imaging (PET) features with anatomical imaging (CT) features. In addition, we removed the reliance on manual input, e.g., tumor delineation, for extracting imaging features.

Main Results: Our experimental results on a well-established benchmark PET-CT dataset show that our method achieved the highest accuracy (0.896) and AUC (0.903) scores when compared to the state-of-the-art methods (unpaired student's t-test p-value < 0.05).

Significance: Our method could be an effective and supportive tool to aid physicians in tumor quantification and in identifying image biomarkers for cancer treatment.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1088/1361-6560/ac3d17DOI Listing
November 2021

Malignant Peritoneal Mesothelioma With EWSR1-ATF1 Fusion: A Case Report.

JTO Clin Res Rep 2021 Nov 11;2(11):100236. Epub 2021 Oct 11.

University of Sydney, Sydney, Australia.

Malignant mesothelioma with EWSR1-ATF1 fusion is a rare malignancy described in young adults without asbestos exposure. To the best of our knowledge, outcomes to local and systemic therapies for this subtype of malignant mesothelioma have not been described. This case report describes the clinical course of a 19-year-old man diagnosed with malignant peritoneal mesothelioma with EWSR1-ATF1 fusion localized to the abdomen. His disease followed an aggressive course and resulted in limited survival (18 mo). There was treatment resistance to several lines of conventional local and systemic treatments for peritoneal mesothelioma and biologically targeted MET inhibition with crizotinib. More research is required in this rare subtype of peritoneal mesothelioma.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.jtocrr.2021.100236DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8569551PMC
November 2021

Diaschisis: a mechanism for subcortical aphasia?

J Neurol 2021 Oct 24. Epub 2021 Oct 24.

Department of Neurology, Royal Prince Alfred Hospital, Sydney, Australia.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s00415-021-10861-7DOI Listing
October 2021

Imaging of patients with multiple myeloma and associated plasma cell disorders: consensus practice statement by the Medical Scientific Advisory Group to Myeloma Australia.

Intern Med J 2021 Oct;51(10):1707-1712

Department of Clinical Haematology, Peter MacCallum Cancer Centre and University of Melbourne, Melbourne, Victoria, Australia.

Imaging modalities for multiple myeloma (MM) have evolved to enable earlier detection of disease. Furthermore, the diagnosis of MM requiring therapy has recently changed to include disease prior to bone destruction, specifically the detection of focal bone lesions. Focal lesions are early, abnormal areas in the bone marrow, which may signal the development of subsequent lytic lesions that typically occur within the next 18-24 months. Cross-sectional imaging modalities are more sensitive for the detection and monitoring of bone and bone marrow disease and are now included in the International Myeloma Working Group current consensus criteria for initial diagnosis and treatment response assessment. The aim of this consensus practice statement is to review the evidence supporting these modalities. A more detailed Position Statement can be found on the Myeloma Australia website.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1111/imj.15457DOI Listing
October 2021

F-FDG PET/CT Radiomics for Preoperative Prediction of Lymph Node Metastases and Nodal Staging in Gastric Cancer.

Front Oncol 2021 13;11:723345. Epub 2021 Sep 13.

Department of Nuclear Medicine, Fudan University Shanghai Cancer Center, Fudan University, Shanghai, China.

Objectives: The accurate assessment of lymph node metastases (LNMs) and the preoperative nodal (N) stage are critical for the precise treatment of patients with gastric cancer (GC). The diagnostic performance, however, of current imaging procedures used for this assessment is sub-optimal. Our aim was to investigate the value of preoperative F-FDG PET/CT radiomic features to predict LNMs and the N stage.

Methods: We retrospectively collected clinical and F-FDG PET/CT imaging data of 185 patients with GC who underwent total or partial radical gastrectomy. Patients were allocated to training and validation sets using the stratified method at a fixed ratio (8:2). There were 2,100 radiomic features extracted from the F-FDG PET/CT scans. After selecting radiomic features by the random forest, relevancy-based, and sequential forward selection methods, the BalancedBagging ensemble classifier was established for the preoperative prediction of LNMs, and the OneVsRest classifier for the N stage. The performance of the models was primarily evaluated by the AUC and accuracy, and validated by the independent validation methods. Analysis of the feature importance and the correlation were also conducted. We also compared the predictive performance of our radiomic models to that with the contrast-enhanced CT (CECT) and F-FDG PET/CT.

Results: There were 185 patients-127 men, 58 women, with the median age of 62, and an age range of 22-86 years. One CT feature and one PET feature were selected to predict LNMs and achieved the best performance (AUC: 82.2%, accuracy: 85.2%). This radiomic model also detected some LNMs that were missed in CECT (19.6%) and F-FDG PET/CT (35.7%). For predicting the N stage, four CT features and one PET feature were selected (AUC: 73.7%, accuracy: 62.3%). Of note, a proportion of patients in the validation set whose LNMs were incorrectly staged by CECT (57.4%) and F-FDG PET/CT (55%) were diagnosed correctly by our radiomic model.

Conclusion: We developed and validated two machine learning models based on the preoperative F-FDG PET/CT images that have a predictive value for LNMs and the N stage in GC. These predictive models show a promise to offer a potentially useful adjunct to current staging approaches for patients with GC.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fonc.2021.723345DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8474469PMC
September 2021

Synthesis and pharmacological evaluation of [F]PBR316: a novel PET ligand targeting the translocator protein 18 kDa (TSPO) with low binding sensitivity to human single nucleotide polymorphism rs6971.

RSC Med Chem 2021 Jul 19;12(7):1207-1221. Epub 2021 Apr 19.

Department of Molecular Imaging, Royal Prince Alfred Hospital Camperdown NSW 2050 Australia

Radiopharmaceuticals that target the translocator protein 18 kDa (TSPO) have been investigated with positron emission tomography (PET) to study neuroinflammation, neurodegeneration and cancer. We have developed the novel, achiral, 2-phenylimidazo[1,2-]pyridine, PBR316 that targets the translocator protein 18 kDa (TSPO) that addresses some of the limitations inherent in current TSPO ligands; namely specificity in binding, blood brain barrier permeability, metabolism and insensitivity to TSPO binding in subjects as a result of rs6971 polymorphism. PBR316 has high nanomolar affinity (4.7-6.0 nM) for the TSPO, >5000 nM for the central benzodiazepine receptor (CBR) and low sensitivity to rs6971 polymorphism with a low affinity binders (LABs) to high affinity binders (HABs) ratio of 1.5. [F]PBR316 was prepared in 20 ± 5% radiochemical yield, >99% radiochemical purity and a molar activity of 160-400 GBq μmol. Biodistribution in rats showed high uptake of [F]PBR316 in organs known to express TSPO such as heart (3.9%) and adrenal glands (7.5% ID per g) at 1 h. [F]PBR316 entered the brain and accumulated in TSPO-expressing regions with an olfactory bulb to brain ratio of 3 at 15 min and 7 at 4 h. Radioactivity was blocked by PK11195 and Ro 5-4864 but not Flumazenil. Metabolite analysis showed that radioactivity in adrenal glands and the brain was predominantly due to the intact radiotracer. PET-CT studies in mouse-bearing prostate tumour xenografts indicated biodistribution similar to rats with radioactivity in the tumour increasing with time. [F]PBR316 shows binding that is insensitive to human polymorphism and has specific and selective binding to the TSPO. [F]PBR316 is suitable for further biological and clinical studies.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1039/d1md00035gDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8292990PMC
July 2021

Pattern and degree of individual brain atrophy predicts dementia onset in dominantly inherited Alzheimer's disease.

Alzheimers Dement (Amst) 2021 5;13(1):e12197. Epub 2021 Jul 5.

The Florey Institute University of Melbourne Parkville Victoria Australia.

Introduction: Asymptomatic and mildly symptomatic dominantly inherited Alzheimer's disease mutation carriers (DIAD-MC) are ideal candidates for preventative treatment trials aimed at delaying or preventing dementia onset. Brain atrophy is an early feature of DIAD-MC and could help predict risk for dementia during trial enrollment.

Methods: We created a dementia risk score by entering standardized gray-matter volumes from 231 DIAD-MC into a logistic regression to classify participants with and without dementia. The score's predictive utility was assessed using Cox models and receiver operating curves on a separate group of 65 DIAD-MC followed longitudinally.

Results: Our risk score separated asymptomatic versus demented DIAD-MC with 96.4% (standard error = 0.02) and predicted conversion to dementia at next visit (hazard ratio = 1.32, 95% confidence interval [CI: 1.15, 1.49]) and within 2 years (area under the curve = 90.3%, 95% CI [82.3%-98.2%]) and improved prediction beyond established methods based on familial age of onset.

Discussion: Individualized risk scores based on brain atrophy could be useful for establishing enrollment criteria and stratifying DIAD-MC participants for prevention trials.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/dad2.12197DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8256623PMC
July 2021

Correction to: Automatic identification of myopic maculopathy related imaging features in optic disc region via machine learning methods.

J Transl Med 2021 May 11;19(1):203. Epub 2021 May 11.

Department of Preventative Ophthalmology, Shanghai Eye Diseases Prevention and Treatment Center, Shanghai Eye Hospital, No. 380 Kangding Road, Shanghai, 200040, China.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12967-021-02874-7DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8114694PMC
May 2021

Automatic identification of myopic maculopathy related imaging features in optic disc region via machine learning methods.

J Transl Med 2021 04 26;19(1):167. Epub 2021 Apr 26.

Department of Preventative Ophthalmology, Shanghai Eye Diseases Prevention and Treatment Center, Shanghai Eye Hospital, No. 380 Kangding Road, Shanghai, 200040, China.

Background: Myopic maculopathy (MM) is the most serious and irreversible complication of pathologic myopia, which is a major cause of visual impairment and blindness. Clinic proposed limited number of factors related to MM. To explore additional features strongly related with MM from optic disc region, we employ a machine learning based radiomics analysis method, which could explore and quantify more hidden or imperceptible MM-related features to the naked eyes and contribute to a more comprehensive understanding of MM and therefore may assist to distinguish the high-risk population in an early stage.

Methods: A total of 457 eyes (313 patients) were enrolled and were divided into severe MM group and without severe MM group. Radiomics analysis was applied to depict features significantly correlated with severe MM from optic disc region. Receiver Operating Characteristic were used to evaluate these features' performance of classifying severe MM.

Results: Eight new MM-related image features were discovered from the optic disc region, which described the shapes, textural patterns and intensity distributions of optic disc region. Compared with clinically reported MM-related features, these newly discovered features exhibited better abilities on severe MM classification. And the mean values of most features were markedly changed between patients with peripapillary diffuse chorioretinal atrophy (PDCA) and macular diffuse chorioretinal atrophy (MDCA).

Conclusions: Machine learning and radiomics method are useful tools for mining more MM-related features from the optic disc region, by which complex or even hidden MM-related features can be discovered and decoded. In this paper, eight new MM-related image features were found, which would be useful for further quantitative study of MM-progression. As a nontrivial byproduct, marked changes between PDCA and MDCA was discovered by both new image features and clinic features.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12967-021-02818-1DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8074495PMC
April 2021

Recurrent feature fusion learning for multi-modality pet-ct tumor segmentation.

Comput Methods Programs Biomed 2021 May 11;203:106043. Epub 2021 Mar 11.

School of Computer Science, University of Sydney, NSW, Australia; Australian Research Council Training Centre for Innovative Bioengineering, NSW, Australia. Electronic address:

Background And Objective: [18f]-fluorodeoxyglucose (fdg) positron emission tomography - computed tomography (pet-ct) is now the preferred imaging modality for staging many cancers. Pet images characterize tumoral glucose metabolism while ct depicts the complementary anatomical localization of the tumor. Automatic tumor segmentation is an important step in image analysis in computer aided diagnosis systems. Recently, fully convolutional networks (fcns), with their ability to leverage annotated datasets and extract image feature representations, have become the state-of-the-art in tumor segmentation. There are limited fcn based methods that support multi-modality images and current methods have primarily focused on the fusion of multi-modality image features at various stages, i.e., early-fusion where the multi-modality image features are fused prior to fcn, late-fusion with the resultant features fused and hyper-fusion where multi-modality image features are fused across multiple image feature scales. Early- and late-fusion methods, however, have inherent, limited freedom to fuse complementary multi-modality image features. The hyper-fusion methods learn different image features across different image feature scales that can result in inaccurate segmentations, in particular, in situations where the tumors have heterogeneous textures.

Methods: we propose a recurrent fusion network (rfn), which consists of multiple recurrent fusion phases to progressively fuse the complementary multi-modality image features with intermediary segmentation results derived at individual recurrent fusion phases: (1) the recurrent fusion phases iteratively learn the image features and then refine the subsequent segmentation results; and, (2) the intermediary segmentation results allows our method to focus on learning the multi-modality image features around these intermediary segmentation results, which minimize the risk of inconsistent feature learning.

Results: we evaluated our method on two pathologically proven non-small cell lung cancer pet-ct datasets. We compared our method to the commonly used fusion methods (early-fusion, late-fusion and hyper-fusion) and the state-of-the-art pet-ct tumor segmentation methods on various network backbones (resnet, densenet and 3d-unet). Our results show that the rfn provides more accurate segmentation compared to the existing methods and is generalizable to different datasets.

Conclusions: we show that learning through multiple recurrent fusion phases allows the iterative re-use of multi-modality image features that refines tumor segmentation results. We also identify that our rfn produces consistent segmentation results across different network architectures.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.cmpb.2021.106043DOI Listing
May 2021

Design, Synthesis, and Biological Evaluation of Novel Fluorescent Probes Targeting the 18-kDa Translocator Protein.

ChemMedChem 2021 Jun 24;16(12):1902-1916. Epub 2021 Mar 24.

School of Chemistry and Molecular Bioscience, and Molecular Horizons, University of Wollongong, Wollongong, NSW, 2522, Australia.

A series of fluorescent probes from the 6-chloro-2-phenylimidazo[1,2-a]pyridine-3-yl acetamides ligands featuring the 7-nitro-2-oxa-1,3-diazol-4-yl (NBD) moiety has been synthesized and biologically evaluated for their fluorescence properties and for their binding affinity to the 18-kDa translocator protein (TSPO). Spectroscopic studies including UV/Vis absorption and fluorescence measurements showed that the synthesized fluorescent probes exhibit favorable spectroscopic properties, especially in nonpolar environments. In vitro fluorescence staining in brain sections from lipopolysaccharide (LPS)-injected mice revealed partial colocalization of the probes with the TSPO. The TSPO binding affinity of the probes was measured on crude mitochondrial fractions separated from rat brain homogenates in a [ C]PK11195 radioligand binding assay. All the new fluorescent probes demonstrated moderate to high binding affinity to the TSPO, with affinity (K ) values ranging from 0.58 nM to 3.28 μM. Taking these data together, we propose that the new fluorescent probes could be used to visualize the TSPO.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/cmdc.202000984DOI Listing
June 2021

Multimodal Spatial Attention Module for Targeting Multimodal PET-CT Lung Tumor Segmentation.

IEEE J Biomed Health Inform 2021 09 3;25(9):3507-3516. Epub 2021 Sep 3.

Multimodal positron emission tomography-computed tomography (PET-CT) is used routinely in the assessment of cancer. PET-CT combines the high sensitivity for tumor detection of PET and anatomical information from CT. Tumor segmentation is a critical element of PET-CT but at present, the performance of existing automated methods for this challenging task is low. Segmentation tends to be done manually by different imaging experts, which is labor-intensive and prone to errors and inconsistency. Previous automated segmentation methods largely focused on fusing information that is extracted separately from the PET and CT modalities, with the underlying assumption that each modality contains complementary information. However, these methods do not fully exploit the high PET tumor sensitivity that can guide the segmentation. We introduce a deep learning-based framework in multimodal PET-CT segmentation with a multimodal spatial attention module (MSAM). The MSAM automatically learns to emphasize regions (spatial areas) related to tumors and suppress normal regions with physiologic high-uptake from the PET input. The resulting spatial attention maps are subsequently employed to target a convolutional neural network (CNN) backbone for segmentation of areas with higher tumor likelihood from the CT image. Our experimental results on two clinical PET-CT datasets of non-small cell lung cancer (NSCLC) and soft tissue sarcoma (STS) validate the effectiveness of our framework in these different cancer types. We show that our MSAM, with a conventional U-Net backbone, surpasses the state-of-the-art lung tumor segmentation approach by a margin of 7.6% in Dice similarity coefficient (DSC).
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/JBHI.2021.3059453DOI Listing
September 2021

Longitudinal Accumulation of Cerebral Microhemorrhages in Dominantly Inherited Alzheimer Disease.

Neurology 2021 03 25;96(12):e1632-e1645. Epub 2021 Jan 25.

From the Departments of Radiology (N.J.-M., T.M.B., B.A.G., G.C., P.M., R.C.H., T.L.S.B.), Neurology (E.M., J.H., B.M.A., R.J.P., J.C.M., R.J.B.), Psychological and Brain Sciences (J.H.), Psychiatry (C.C., C.M.K.), and Pathology and Immunology (R.J.P.) and Division of Biostatistics (G.W., C.X.), Washington University School of Medicine, St. Louis, MO; Banner Alzheimers Institute (Y.S.), Phoenix, AZ; Department of Cognitive Neurology and Neuropsychology (R.F.A.), Instituto de Investigaciones Neurológicas Fleni, Buenos Aires, Argentina; Departments of Neurology and Clinical and Translational Science (S.B.B.), University of Pittsburgh School of Medicine, PA; Department of Neurology (A.M.B.), Taub Institute for Research on Alzheimers Disease and the Aging Brain, College of Physicians and Surgeons, Columbia University, New York, NY; Neuroscience Research Australia (W.S.B., P.R.S.); School of Medical Sciences (P.R.S.), University of New South Wales (W.S.B.), Sydney, Australia; Dementia Research Centre and UK Dementia Research Institute (D.M.C., N.C.F., A.O.), UCL Queen Square Institute of Neurology, London, UK; Departments of Neurology (J.P.C., K.A.J.) and Radiology (K.A.J.), Massachusetts General Hospital, Boston; Department of Neurology (H.C.C., J.M.R.), Keck School of Medicine of USC, Los Angeles, CA; Department of Psychiatry and Human Behavior (S.C., A.K.W.L., S.S.), Memory and Aging Program, Butler Hospital, Brown University Alpert Medical School, Providence, RI; Center for Neuroimaging, Department of Radiology and Imaging Science (M.R.F., A.J.S.), Department of Pathology and Laboratory Medicine (B.G.), and Indiana Alzheimers Disease Research Center (A.J.S.), Indiana University School of Medicine, Indianapolis; Departments of Molecular Imaging and Neurology (M.F.), Royal Prince Alfred Hospital, University of Sydney, Australia; Department of Neurology (N.R.G.-R.), Mayo Clinic, Jacksonville, FL; German Center for Neurodegenerative Diseases (DZNE) (C.L., J.L., I.Y.); Section for Dementia Research, Hertie Institute for Clinical Brain Research and Department of Psychiatry and Psychotherapy (C.L.), University of Tübingen; Department of Neurology (J.L., I.Y.), Ludwig-Maximilians-Universität München; Munich Cluster for Systems Neurology (SyNergy) (J.L., I.Y.), Germany; Florey Institute and The University of Melbourne (C.L.M.), Australia; Department of Neurology (J.M.N.), Columbia University Irving Medical Center, New York, NY; Department of Radiology (K.K., C.R.J., G.M.P.), Mayo Clinic, Rochester, MN; Department of Molecular Imaging and Therapy (C.C.R., V.L.V.), Austin Health, University of Melbourne, Heidelberg, Australia; Clinical Research Center for Dementia (H.S.), Osaka City University; Department of Neurology (M.S.), Hirosaki University Graduate School of Medicine; and Department of Neurology (K.S.), The University of Tokyo, Japan.

Objective: To investigate the inherent clinical risks associated with the presence of cerebral microhemorrhages (CMHs) or cerebral microbleeds and characterize individuals at high risk for developing hemorrhagic amyloid-related imaging abnormality (ARIA-H), we longitudinally evaluated families with dominantly inherited Alzheimer disease (DIAD).

Methods: Mutation carriers (n = 310) and noncarriers (n = 201) underwent neuroimaging, including gradient echo MRI sequences to detect CMHs, and neuropsychological and clinical assessments. Cross-sectional and longitudinal analyses evaluated relationships between CMHs and neuroimaging and clinical markers of disease.

Results: Three percent of noncarriers and 8% of carriers developed CMHs primarily located in lobar areas. Carriers with CMHs were older, had higher diastolic blood pressure and Hachinski ischemic scores, and more clinical, cognitive, and motor impairments than those without CMHs. ε4 status was not associated with the prevalence or incidence of CMHs. Prevalent or incident CMHs predicted faster change in Clinical Dementia Rating although not composite cognitive measure, cortical thickness, hippocampal volume, or white matter lesions. Critically, the presence of 2 or more CMHs was associated with a significant risk for development of additional CMHs over time (8.95 ± 10.04 per year).

Conclusion: Our study highlights factors associated with the development of CMHs in individuals with DIAD. CMHs are a part of the underlying disease process in DIAD and are significantly associated with dementia. This highlights that in participants in treatment trials exposed to drugs, which carry the risk of ARIA-H as a complication, it may be challenging to separate natural incidence of CMHs from drug-related CMHs.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1212/WNL.0000000000011542DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8032370PMC
March 2021

Modeling autosomal dominant Alzheimer's disease with machine learning.

Alzheimers Dement 2021 06 21;17(6):1005-1016. Epub 2021 Jan 21.

German Center for Neurodegenerative Diseases, Munich, Germany.

Introduction: Machine learning models were used to discover novel disease trajectories for autosomal dominant Alzheimer's disease.

Methods: Longitudinal structural magnetic resonance imaging, amyloid positron emission tomography (PET), and fluorodeoxyglucose PET were acquired in 131 mutation carriers and 74 non-carriers from the Dominantly Inherited Alzheimer Network; the groups were matched for age, education, sex, and apolipoprotein ε4 (APOE ε4). A deep neural network was trained to predict disease progression for each modality. Relief algorithms identified the strongest predictors of mutation status.

Results: The Relief algorithm identified the caudate, cingulate, and precuneus as the strongest predictors among all modalities. The model yielded accurate results for predicting future Pittsburgh compound B (R  = 0.95), fluorodeoxyglucose (R  = 0.93), and atrophy (R  = 0.95) in mutation carriers compared to non-carriers.

Discussion: Results suggest a sigmoidal trajectory for amyloid, a biphasic response for metabolism, and a gradual decrease in volume, with disease progression primarily in subcortical, middle frontal, and posterior parietal regions.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/alz.12259DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8195816PMC
June 2021

Comparing cortical signatures of atrophy between late-onset and autosomal dominant Alzheimer disease.

Neuroimage Clin 2020 5;28:102491. Epub 2020 Nov 5.

Department of Radiology, Department of Neurology, Department of Psychiatry, Department of Pathology and Immunology, Division of Biostatistics, Washington University School of Medicine, Saint Louis, MO, USA.

Defining a signature of cortical regions of interest preferentially affected by Alzheimer disease (AD) pathology may offer improved sensitivity to early AD compared to hippocampal volume or mesial temporal lobe alone. Since late-onset Alzheimer disease (LOAD) participants tend to have age-related comorbidities, the younger-onset age in autosomal dominant AD (ADAD) may provide a more idealized model of cortical thinning in AD. To test this, the goals of this study were to compare the degree of overlap between the ADAD and LOAD cortical thinning maps and to evaluate the ability of the ADAD cortical signature regions to predict early pathological changes in cognitively normal individuals. We defined and analyzed the LOAD cortical maps of cortical thickness in 588 participants from the Knight Alzheimer Disease Research Center (Knight ADRC) and the ADAD cortical maps in 269 participants from the Dominantly Inherited Alzheimer Network (DIAN) observational study. Both cohorts were divided into three groups: cognitively normal controls (n = 381; n = 145), preclinical (n = 153; n = 76), and cognitively impaired (n = 54; n = 48). Both cohorts underwent clinical assessments, 3T MRI, and amyloid PET imaging with either C-Pittsburgh compound B or F-florbetapir. To generate cortical signature maps of cortical thickness, we performed a vertex-wise analysis between the cognitively normal controls and impaired groups within each cohort using six increasingly conservative statistical thresholds to determine significance. The optimal cortical map among the six statistical thresholds was determined from a receiver operating characteristic analysis testing the performance of each map in discriminating between the cognitively normal controls and preclinical groups. We then performed within-cohort and cross-cohort (e.g. ADAD maps evaluated in the Knight ADRC cohort) analyses to examine the sensitivity of the optimal cortical signature maps to the amyloid levels using only the cognitively normal individuals (cognitively normal controls and preclinical groups) in comparison to hippocampal volume. We found the optimal cortical signature maps were sensitive to early increases in amyloid for the asymptomatic individuals within their respective cohorts and were significant beyond the inclusion of hippocampus volume, but the cortical signature maps performed poorly when analyzing across cohorts. These results suggest the cortical signature maps are a useful MRI biomarker of early AD-related neurodegeneration in preclinical individuals and the pattern of decline differs between LOAD and ADAD.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.nicl.2020.102491DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7689410PMC
June 2021

An Intraocular Thymic Metastasis Identified on 18F-FDG PET/CT Before and After Treatment.

Clin Nucl Med 2021 Mar;46(3):240-242

From the Department of Molecular Imaging, Royal Prince Alfred Hospital, Camperdown.

Abstract: We present the imaging findings pretreatment and posttreatment in a 58-year-old woman with recurrent thymic carcinoma. Two years after treatment, the patient presented with a 3-week history of right eye pain and blurred vision. Ophthalmological examination and MRI of the orbits showed a right superolateral choroidal lesion. Neurologic and whole-body FDG PET/CT scans showed a markedly glucose-avid right choroidal mass and extensive lung parenchymal, pleural, and thoracic nodal disease. There was a good response to chemoradiotherapy with a reduction in size and metabolism at all sites.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1097/RLU.0000000000003431DOI Listing
March 2021

Cyclotron-based production of Ga, [Ga]GaCl, and [Ga]Ga-PSMA-11 from a liquid target.

EJNMMI Radiopharm Chem 2020 Nov 12;5(1):25. Epub 2020 Nov 12.

Division of Nuclear Medicine, Department of Radiology, University of Michigan, Ann Arbor, MI, USA.

Purpose: To optimize the direct production of Ga on a cyclotron, via the Zn(p,n)Ga reaction using a liquid cyclotron target. We Investigated the yield of cyclotron-produced Ga, extraction of [Ga]GaCl and subsequent [Ga]Ga-PSMA-11 labeling using an automated synthesis module.

Methods: Irradiations of a 1.0 M solution of [Zn]Zn(NO) in dilute (0.2-0.3 M) HNO were conducted using GE PETtrace cyclotrons and GE Ga liquid targets. The proton beam energy was degraded to a nominal 14.3 MeV to minimize the co-production of Ga through the Zn(p,2n)Ga reaction without unduly compromising Ga yields. We also evaluated the effects of varying beam times (50-75 min) and beam currents (27-40 μA). Crude Ga production was measured. The extraction of [Ga]GaCl was performed using a 2 column solid phase method on the GE FASTlab Developer platform. Extracted [Ga]GaCl was used to label [Ga]Ga-PSMA-11 that was intended for clinical use.

Results: The decay corrected yield of Ga at EOB was typically > 3.7 GBq (100 mCi) for a 60 min beam, with irradiations of [Zn]Zn(NO) at 0.3 M HNO Target/chemistry performance was more consistent when compared with 0.2 M HNO. Radionuclidic purity of Ga was typically > 99.8% at EOB and met the requirements specified in the European Pharmacopoeia (< 2% combined Ga) for a practical clinical product shelf-life. The activity yield of [Ga]GaCl was typically > 50% (~ 1.85 GBq, 50 mCi); yields improved as processes were optimized. Labeling yields for [Ga]Ga-PSMA-11 were near quantitative (~ 1.67 GBq, 45 mCi) at EOS. Cyclotron produced [Ga]Ga-PSMA-11 underwent full quality control, stability and sterility testing, and was implemented for human use at the University of Michigan as an Investigational New Drug through the US FDA and also at the Royal Prince Alfred Hospital (RPA).

Conclusion: Direct cyclotron irradiation of a liquid target provides clinically relevant quantities of [Ga]Ga-PSMA-11 and is a viable alternative to traditional Ge/Ga generators.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s41181-020-00106-9DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7661618PMC
November 2020

Unsupervised Domain Adaptation to Classify Medical Images Using Zero-Bias Convolutional Auto-Encoders and Context-Based Feature Augmentation.

IEEE Trans Med Imaging 2020 07 3;39(7):2385-2394. Epub 2020 Feb 3.

The accuracy and robustness of image classification with supervised deep learning are dependent on the availability of large-scale labelled training data. In medical imaging, these large labelled datasets are sparse, mainly related to the complexity in manual annotation. Deep convolutional neural networks (CNNs), with transferable knowledge, have been employed as a solution to limited annotated data through: 1) fine-tuning generic knowledge with a relatively smaller amount of labelled medical imaging data, and 2) learning image representation that is invariant to different domains. These approaches, however, are still reliant on labelled medical image data. Our aim is to use a new hierarchical unsupervised feature extractor to reduce reliance on annotated training data. Our unsupervised approach uses a multi-layer zero-bias convolutional auto-encoder that constrains the transformation of generic features from a pre-trained CNN (for natural images) to non-redundant and locally relevant features for the medical image data. We also propose a context-based feature augmentation scheme to improve the discriminative power of the feature representation. We evaluated our approach on 3 public medical image datasets and compared it to other state-of-the-art supervised CNNs. Our unsupervised approach achieved better accuracy when compared to other conventional unsupervised methods and baseline fine-tuned CNNs.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2020.2971258DOI Listing
July 2020

F-FDG PET/CT radiomic predictors of pathologic complete response (pCR) to neoadjuvant chemotherapy in breast cancer patients.

Eur J Nucl Med Mol Imaging 2020 05 25;47(5):1116-1126. Epub 2020 Jan 25.

Shanghai Key Laboratory for Molecular Imaging, Shanghai University of Medicine and Health Sciences, Shanghai, China.

Purpose: Pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) is commonly accepted as the gold standard to assess outcome after NAC in breast cancer patients. F-Fluorodeoxyglucose positron emission tomography/computed tomography (PET/CT) has unique value in tumor staging, predicting prognosis, and evaluating treatment response. Our aim was to determine if we could identify radiomic predictors from PET/CT in breast cancer patient therapeutic efficacy prior to NAC.

Methods: This retrospective study included 100 breast cancer patients who received NAC; there were 2210 PET/CT radiomic features extracted. Unsupervised and supervised machine learning models were used to identify the prognostic radiomic predictors through the following: (1) selection of the significant (p < 0.05) imaging features from consensus clustering and the Wilcoxon signed-rank test; (2) selection of the most discriminative features via univariate random forest (Uni-RF) and the Pearson correlation matrix (PCM); and (3) determination of the most predictive features from a traversal feature selection (TFS) based on a multivariate random forest (RF). The prediction model was constructed with RF and then validated with 10-fold cross-validation for 30 times and then independently validated. The performance of the radiomic predictors was measured in terms of area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

Results: The PET/CT radiomic predictors achieved a prediction accuracy of 0.857 (AUC = 0.844) on the training split set and 0.767 (AUC = 0.722) on the independent validation set. When age was incorporated, the accuracy for the split set increased to 0.857 (AUC = 0.958) and 0.8 (AUC = 0.73) for the independent validation set and both outperformed the clinical prediction model. We also found a close association between the radiomic features, receptor expression, and tumor T stage.

Conclusion: Radiomic predictors from pre-treatment PET/CT scans when combined with patient age were able to predict pCR after NAC. We suggest that these data will be valuable for patient management.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s00259-020-04684-3DOI Listing
May 2020

Deep multi-modality collaborative learning for distant metastases predication in PET-CT soft-tissue sarcoma studies.

Annu Int Conf IEEE Eng Med Biol Soc 2019 Jul;2019:3658-3688

Soft-tissue Sarcomas (STS) are a heterogeneous group of malignant neoplasms with a relatively high mortality rate from distant metastases. Early prediction or quantitative evaluation of distant metastases risk for patients with STS is an important step which can provide better-personalized treatments and thereby improve survival rates. Positron emission tomography-computed tomography (PET-CT) image is regarded as the imaging modality of choice for the evaluation, staging and assessment of STS. Radiomics, which refers to the extraction and analysis of the quantitative of high-dimensional mineable data from medical images, is foreseen as an important prognostic tool for cancer risk assessment. However, conventional radiomics methods that depend heavily on hand-crafted features (e.g. shape and texture) and prior knowledge (e.g. tuning of many parameters) therefore cannot fully represent the semantic information of the image. In addition, convolutional neural networks (CNN) based radiomics methods present capabilities to improve, but currently, they are mainly designed for single modality e.g., CT or a particular body region e.g., lung structure. In this work, we propose a deep multi-modality collaborative learning to iteratively derive optimal ensembled deep and conventional features from PET-CT images. In addition, we introduce an end-to-end volumetric deep learning architecture to learn complementary PET-CT features optimised for image radiomics. Our experimental results using public PET-CT dataset of STS patients demonstrate that our method has better performance when compared with the state-of-the-art methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC.2019.8857666DOI Listing
July 2019

Convolutional sparse kernel network for unsupervised medical image analysis.

Med Image Anal 2019 08 12;56:140-151. Epub 2019 Jun 12.

School of Computer Science, University of Sydney, NSW, Australia. Electronic address:

The availability of large-scale annotated image datasets and recent advances in supervised deep learning methods enable the end-to-end derivation of representative image features that can impact a variety of image analysis problems. Such supervised approaches, however, are difficult to implement in the medical domain where large volumes of labelled data are difficult to obtain due to the complexity of manual annotation and inter- and intra-observer variability in label assignment. We propose a new convolutional sparse kernel network (CSKN), which is a hierarchical unsupervised feature learning framework that addresses the challenge of learning representative visual features in medical image analysis domains where there is a lack of annotated training data. Our framework has three contributions: (i) we extend kernel learning to identify and represent invariant features across image sub-patches in an unsupervised manner. (ii) We initialise our kernel learning with a layer-wise pre-training scheme that leverages the sparsity inherent in medical images to extract initial discriminative features. (iii) We adapt a multi-scale spatial pyramid pooling (SPP) framework to capture subtle geometric differences between learned visual features. We evaluated our framework in medical image retrieval and classification on three public datasets. Our results show that our CSKN had better accuracy when compared to other conventional unsupervised methods and comparable accuracy to methods that used state-of-the-art supervised convolutional neural networks (CNNs). Our findings indicate that our unsupervised CSKN provides an opportunity to leverage unannotated big data in medical imaging repositories.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.media.2019.06.005DOI Listing
August 2019

Co-Learning Feature Fusion Maps from PET-CT Images of Lung Cancer.

IEEE Trans Med Imaging 2019 Jun 17. Epub 2019 Jun 17.

The analysis of multi-modality positron emission tomography and computed tomography (PET-CT) images for computer aided diagnosis applications (e.g., detection and segmentation) requires combining the sensitivity of PET to detect abnormal regions with anatomical localization from CT. Current methods for PET-CT image analysis either process the modalities separately or fuse information from each modality based on knowledge about the image analysis task. These methods generally do not consider the spatially varying visual characteristics that encode different information across the different modalities, which have different priorities at different locations. For example, a high abnormal PET uptake in the lungs is more meaningful for tumor detection than physiological PET uptake in the heart. Our aim is to improve fusion of the complementary information in multi-modality PET-CT with a new supervised convolutional neural network (CNN) that learns to fuse complementary information for multi-modality medical image analysis. Our CNN first encodes modality-specific features and then uses them to derive a spatially varying fusion map that quantifies the relative importance of each modality's features across different spatial locations. These fusion maps are then multiplied with the modality-specific feature maps to obtain a representation of the complementary multi-modality information at different locations, which can then be used for image analysis. We evaluated the ability of our CNN to detect and segment multiple regions (lungs, mediastinum, tumors) with different fusion requirements using a dataset of PET-CT images of lung cancer. We compared our method to baseline techniques for multi-modality image fusion (fused inputs (FS), multi-branch (MB) techniques, and multichannel (MC) techniques) and segmentation. Our findings show that our CNN had a significantly higher foreground detection accuracy (99.29%, p < 0:05) than the fusion baselines (FS: 99.00%, MB: 99.08%, TC: 98.92%) and a significantly higher Dice score (63.85%) than recent PET-CT tumor segmentation methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2019.2923601DOI Listing
June 2019

A web-based multidisciplinary team meeting visualisation system.

Int J Comput Assist Radiol Surg 2019 Dec 21;14(12):2221-2231. Epub 2019 May 21.

Biomedical and Multimedia Information Technology Research Group, School of Computer Science, The University of Sydney, Sydney, Australia.

Purpose: Multidisciplinary team meetings (MDTs) are the standard of care for safe, effective patient management in modern hospital-based clinical practice. Medical imaging data are often the central discussion points in many MDTs, and these data are typically visualised, by all participants, on a common large display. We propose a Web-based MDT visualisation system (WMDT-VS) to allow individual participants to view the data on their own personal computing devices with the potential to customise the imaging data, i.e. different view of the data to that of the common display, for their particular clinical perspective.

Methods: We developed the WMDT-VS by leveraging the state-of-the-art Web technologies to support four MDT visualisation features: (1) 2D and 3D visualisations for multiple imaging modality data; (2) a variety of personal computing devices, e.g. smartphone, tablets, laptops and PCs, to access and navigate medical images individually and share the visualisations; (3) customised participant visualisations; and (4) the addition of extra local image data for visualisation and discussion.

Results: We outlined these MDT visualisation features on two simulated MDT settings using different imaging data and usage scenarios. We measured compatibility and performances of various personal, consumer-level, computing devices.

Conclusions: Our WMDT-VS provides a more comprehensive visualisation experience for MDT participants.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s11548-019-01999-xDOI Listing
December 2019

Comparison of Pittsburgh compound B and florbetapir in cross-sectional and longitudinal studies.

Alzheimers Dement (Amst) 2019 Dec 22;11:180-190. Epub 2019 Feb 22.

The University of Melbourne and the Florey Institute, Parkville, VIC, Australia.

Introduction: Quantitative measurement of brain amyloid burden is important for both research and clinical purposes. However, the existence of multiple imaging tracers presents challenges to the interpretation of such measurements. This study presents a direct comparison of Pittsburgh compound B-based and florbetapir-based amyloid imaging in the same participants from two independent cohorts using a crossover design.

Methods: Pittsburgh compound B and florbetapir amyloid PET imaging data from three different cohorts were analyzed using previously established pipelines to obtain global amyloid burden measurements. These measurements were converted to the Centiloid scale to allow fair comparison between the two tracers. The mean and inter-individual variability of the two tracers were compared using multivariate linear models both cross-sectionally and longitudinally.

Results: Global amyloid burden measured using the two tracers were strongly correlated in both cohorts. However, higher variability was observed when florbetapir was used as the imaging tracer. The variability may be partially caused by white matter signal as partial volume correction reduces the variability and improves the correlations between the two tracers. Amyloid burden measured using both tracers was found to be in association with clinical and psychometric measurements. Longitudinal comparison of the two tracers was also performed in similar but separate cohorts whose baseline amyloid load was considered elevated (i.e., amyloid positive). No significant difference was detected in the average annualized rate of change measurements made with these two tracers.

Discussion: Although the amyloid burden measurements were quite similar using these two tracers as expected, difference was observable even after conversion into the Centiloid scale. Further investigation is warranted to identify optimal strategies to harmonize amyloid imaging data acquired using different tracers.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.dadm.2018.12.008DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6389727PMC
December 2019

A propagation-DNN: Deep combination learning of multi-level features for MR prostate segmentation.

Comput Methods Programs Biomed 2019 Mar 29;170:11-21. Epub 2018 Dec 29.

Biomedical and Multimedia Information Technology Research Group, School of Computer Science, University of Sydney, Sydney, Australia.

Background And Objective: Prostate segmentation on Magnetic Resonance (MR) imaging is problematic because disease changes the shape and boundaries of the gland and it can be difficult to separate the prostate from surrounding tissues. We propose an automated model that extracts and combines multi-level features in a deep neural network to segment prostate on MR images.

Methods: Our proposed model, the Propagation Deep Neural Network (P-DNN), incorporates the optimal combination of multi-level feature extraction as a single model. High level features from the convolved data using DNN are extracted for prostate localization and shape recognition, while labeling propagation, by low level cues, is embedded into a deep layer to delineate the prostate boundary.

Results: A well-recognized benchmarking dataset (50 training data and 30 testing data from patients) was used to evaluate the P-DNN. When compared it to existing DNN methods, the P-DNN statistically outperformed the baseline DNN models with an average improvement in the DSC of 3.19%. When compared to the state-of-the-art non-DNN prostate segmentation methods, P-DNN was competitive by achieving 89.9 ± 2.8% DSC and 6.84 ± 2.5 mm HD on training sets and 84.13 ± 5.18% DSC and 9.74 ± 4.21 mm HD on testing sets.

Conclusion: Our results show that P-DNN maximizes multi-level feature extraction for prostate segmentation of MR images.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.cmpb.2018.12.031DOI Listing
March 2019

A direct volume rendering visualization approach for serial PET-CT scans that preserves anatomical consistency.

Int J Comput Assist Radiol Surg 2019 May 19;14(5):733-744. Epub 2019 Jan 19.

Sydney Medical School, The University of Sydney, Sydney, Australia.

Purpose: Our aim was to develop an interactive 3D direct volume rendering (DVR) visualization solution to interpret and analyze complex, serial multi-modality imaging datasets from positron emission tomography-computed tomography (PET-CT).

Methods: Our approach uses: (i) a serial transfer function (TF) optimization to automatically depict particular regions of interest (ROIs) over serial datasets with consistent anatomical structures; (ii) integration of a serial segmentation algorithm to interactively identify and track ROIs on PET; and (iii) parallel graphics processing unit (GPU) implementation for interactive visualization.

Results: Our DVR visualization more easily identifies changes in ROIs in serial scans in an automated fashion and parallel GPU computation which enables interactive visualization.

Conclusions: Our approach provides a rapid 3D visualization of relevant ROIs over multiple scans, and we suggest that it can be used as an adjunct to conventional 2D viewing software from scanner vendors.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s11548-019-01916-2DOI Listing
May 2019

Unsupervised Two-Path Neural Network for Cell Event Detection and Classification Using Spatiotemporal Patterns.

IEEE Trans Med Imaging 2019 06 7;38(6):1477-1487. Epub 2018 Dec 7.

Automatic event detection in cell videos is essential for monitoring cell populations in biomedicine. Deep learning methods have advantages over traditional approaches for cell event detection due to their ability to capture more discriminative features of cellular processes. Supervised deep learning methods, however, are inherently limited due to the scarcity of annotated data. Unsupervised deep learning methods have shown promise in general (non-cell) videos because they can learn the visual appearance and motion of regularly occurring events. Cell videos, however, can have rapid, irregular changes in cell appearance and motion, such as during cell division and death, which are often the events of most interest. We propose a novel unsupervised two-path input neural network architecture to capture these irregular events with three key elements: 1) a visual encoding path to capture regular spatiotemporal patterns of observed objects with convolutional long short-term memory units; 2) an event detection path to extract information related to irregular events with max-pooling layers; and 3) integration of the hidden states of the two paths to provide a comprehensive representation of the video that is used to simultaneously locate and classify cell events. We evaluated our network in detecting cell division in densely packed stem cells in phase-contrast microscopy videos. Our unsupervised method achieved higher or comparable accuracy to standard and state-of-the-art supervised methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2018.2885572DOI Listing
June 2019

NMDA receptor antibody in teratoma-related opsoclonus-myoclonus syndrome.

J Clin Neurosci 2018 Dec 16;58:203-204. Epub 2018 Oct 16.

Department of Immunology, Royal Prince Alfred Hospital, Sydney, Australia.

Opsoclonus-myoclonus syndrome (OMS) is a brainstem/cerebellar syndrome producing disabling multi-directional saccadic oscillations with oscillopsia, with or without somatic myoclonus and cerebellar ataxia (Wong et al., 2001; Armangué et al., 2016). OMS is presumed to have an autoimmune basis and patients with it are tested for antineuronal antibodies and have imaging to locate any tumors. Here we report a unusual case of a young woman who had NMDAR antibody (NMDAR-ab) positive, teratoma-related, isolated OMS without encephalopathy. Removal of her ovarian teratoma, and immunotherapy with steroids, intravenous immunoglobulin (IVIg), plasma exchange (PLEX), and ultimately with B-cell depletion with rituximab resulted in total recovery after 3 months. Patients with teratoma-related OMS very rarely have NMDAR-ab which suggests that it is not the NMDAR-ab per se that causes the OMS.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.jocn.2018.10.011DOI Listing
December 2018

Knowledge-based Collaborative Deep Learning for Benign-Malignant Lung Nodule Classification on Chest CT.

IEEE Trans Med Imaging 2019 04 17;38(4):991-1004. Epub 2018 Oct 17.

The accurate identification of malignant lung nodules on chest CT is critical for the early detection of lung cancer, which also offers patients the best chance of cure. Deep learning methods have recently been successfully introduced to computer vision problems, although substantial challenges remain in the detection of malignant nodules due to the lack of large training data sets. In this paper, we propose a multi-view knowledge-based collaborative (MV-KBC) deep model to separate malignant from benign nodules using limited chest CT data. Our model learns 3-D lung nodule characteristics by decomposing a 3-D nodule into nine fixed views. For each view, we construct a knowledge-based collaborative (KBC) submodel, where three types of image patches are designed to fine-tune three pre-trained ResNet-50 networks that characterize the nodules' overall appearance, voxel, and shape heterogeneity, respectively. We jointly use the nine KBC submodels to classify lung nodules with an adaptive weighting scheme learned during the error back propagation, which enables the MV-KBC model to be trained in an end-to-end manner. The penalty loss function is used for better reduction of the false negative rate with a minimal effect on the overall performance of the MV-KBC model. We tested our method on the benchmark LIDC-IDRI data set and compared it to the five state-of-the-art classification approaches. Our results show that the MV-KBC model achieved an accuracy of 91.60% for lung nodule classification with an AUC of 95.70%. These results are markedly superior to the state-of-the-art approaches.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2018.2876510DOI Listing
April 2019

Rheumatoid leptomeningitis presenting with an acute neuropsychiatric disorder.

Pract Neurol 2019 Feb 10;19(1):68-71. Epub 2018 Aug 10.

Department of Neurology, Royal Prince Alfred Hospital, Camperdown, New South Wales, Australia.

Leptomeningitis is a rare central nervous system manifestation of rheumatoid arthritis, generally in patients with established chronic rheumatoid disease. We report a 41-year-old man without previous rheumatoid arthritis or psychiatric disorder who presented with an acute neuropsychiatric disturbance and polyarthralgia. His MR scan of brain showed asymmetric bifrontal leptomeningitis, confirmed on (18F)-fluoro-D-glucose-positron emission tomography. Other investigations showed highly positive serum and cerebrospinal fluid anti-cyclic citrullinated peptide. A leptomeningeal biopsy showed necrotising leptomeningeal inflammation with ill-defined granulomas and lymphoplasmacytic infiltrate without organisms. Prolonged high-dose corticosteroids and then rituximab resulted in recovery. Chronic leptomeningitis can present with an acute neuropsychiatric disorder. We highlight that early rheumatoid disease can, rarely, cause a chronic leptomeningitis, reversible with immunotherapy.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1136/practneurol-2018-001978DOI Listing
February 2019
-->