Publications by authors named "Jay Engel"

7 Publications

  • Page 1 of 1

Domain adaptation and self-supervised learning for surgical margin detection.

Int J Comput Assist Radiol Surg 2021 May 6. Epub 2021 May 6.

School of Computing, Queen's University, Ontario, Canada.

Purpose: One in five women who undergo breast conserving surgery will need a second revision surgery due to remaining tumor. The iKnife is a mass spectrometry modality that produces real-time margin information based on the metabolite signatures in surgical smoke. Using this modality and real-time tissue classification, surgeons could remove all cancerous tissue during the initial surgery, improving many facets of patient outcomes. An obstacle in developing a iKnife breast cancer recognition model is the destructive, time-consuming and sensitive nature of the data collection that limits the size of the datasets.

Methods: We address these challenges by first, building a self-supervised learning model from limited, weakly labeled data. By doing so, the model can learn to contextualize the general features of iKnife data from a more accessible cancer type. Second, the trained model can then be applied to a cancer classification task on breast data. This domain adaptation allows for the transfer of learnt weights from models of one tissue type to another.

Results: Our datasets contained 320 skin burns (129 tumor burns, 191 normal burns) from 51 patients and 144 breast tissue burns (41 tumor and 103 normal) from 11 patients. We investigate the effect of different hyper-parameters on the performance of the final classifier. The proposed two-step method performed statistically significantly better than a baseline model (p-value < 0.0001), by achieving an accuracy, sensitivity and specificity of 92%, 88% and 92%, respectively.

Conclusion: This is the first application of domain transfer for iKnife REIMS data. We showed that having a limited number of breast data samples for training a classifier can be compensated by self-supervised learning and domain adaption on a set of unlabeled skin data. We plan to confirm this performance by collecting new breast samples and extending it to incorporate other cancer tissues.
View Article and Find Full Text PDF

Download full-text PDF

Source Listing
May 2021

Comprehensive Molecular Characterization of Papillary Renal-Cell Carcinoma.

N Engl J Med 2016 Jan 4;374(2):135-45. Epub 2015 Nov 4.

Background: Papillary renal-cell carcinoma, which accounts for 15 to 20% of renal-cell carcinomas, is a heterogeneous disease that consists of various types of renal cancer, including tumors with indolent, multifocal presentation and solitary tumors with an aggressive, highly lethal phenotype. Little is known about the genetic basis of sporadic papillary renal-cell carcinoma, and no effective forms of therapy for advanced disease exist.

Methods: We performed comprehensive molecular characterization of 161 primary papillary renal-cell carcinomas, using whole-exome sequencing, copy-number analysis, messenger RNA and microRNA sequencing, DNA-methylation analysis, and proteomic analysis.

Results: Type 1 and type 2 papillary renal-cell carcinomas were shown to be different types of renal cancer characterized by specific genetic alterations, with type 2 further classified into three individual subgroups on the basis of molecular differences associated with patient survival. Type 1 tumors were associated with MET alterations, whereas type 2 tumors were characterized by CDKN2A silencing, SETD2 mutations, TFE3 fusions, and increased expression of the NRF2-antioxidant response element (ARE) pathway. A CpG island methylator phenotype (CIMP) was observed in a distinct subgroup of type 2 papillary renal-cell carcinomas that was characterized by poor survival and mutation of the gene encoding fumarate hydratase (FH).

Conclusions: Type 1 and type 2 papillary renal-cell carcinomas were shown to be clinically and biologically distinct. Alterations in the MET pathway were associated with type 1, and activation of the NRF2-ARE pathway was associated with type 2; CDKN2A loss and CIMP in type 2 conveyed a poor prognosis. Furthermore, type 2 papillary renal-cell carcinoma consisted of at least three subtypes based on molecular and phenotypic features. (Funded by the National Institutes of Health.).
View Article and Find Full Text PDF

Download full-text PDF

Source Listing
January 2016

Navigated Breast Tumor Excision Using Electromagnetically Tracked Ultrasound and Surgical Instruments.

IEEE Trans Biomed Eng 2016 Mar 20;63(3):600-6. Epub 2015 Aug 20.

Objective: Lumpectomy, breast conserving tumor excision, is the standard surgical treatment in early stage breast cancer. A common problem with lumpectomy is that the tumor may not be completely excised, and additional surgery becomes necessary. We investigated if a surgical navigation system using intraoperative ultrasound improves the outcomes of lumpectomy and if such a system can be implemented in the clinical environment.

Methods: Position sensors were applied on the tumor localization needle, the ultrasound probe, and the cautery, and 3-D navigation views were generated using real-time tracking information. The system was tested against standard wire-localization procedures on phantom breast models by eight surgical residents. Clinical safety and feasibility was tested in six palpable tumor patients undergoing lumpectomy by two experienced surgical oncologists.

Results: Navigation resulted in significantly less tissue excised compared to control procedures (10.3 ± 4.4 versus 18.6 ± 8.7 g, p = 0.01) and lower number of tumor-positive margins (1/8 versus 4/8) in the phantom experiments. Excision-tumor distance was also more consistently outside the tumor margins with navigation in phantoms. The navigation system has been successfully integrated in an operating room, and user experience was rated positively by surgical oncologists.

Conclusion: Electromagnetic navigation may improve the outcomes of lumpectomy by making the tumor excision more accurate.

Significance: Breast cancer is the most common cancer in women, and lumpectomy is its first choice treatment. Therefore, the improvement of lumpectomy outcomes has a significant impact on a large patient population.
View Article and Find Full Text PDF

Download full-text PDF

Source Listing
March 2016

Role of the integrin-binding protein osteopontin in lymphatic metastasis of breast cancer.

Am J Pathol 2006 Jul;169(1):233-46

Department of Oncology, Schulich School of Medicine and Dentistry, University of Western Ontario, London Regional Cancer Program, 790 Commissioners Road East, London, Ontario N6A 4L6, Canada.

Although a primary route of breast cancer metastasis is believed to be via lymphatics, the molecular factors involved are poorly understood. We hypothesized that one such factor may be the integrin-binding protein osteopontin (OPN), and we investigated this clinically and experimentally. In breast cancer patients undergoing sentinel lymph node biopsy, OPN levels were significantly higher in lymph node metastases than in the primary tumor (P < 0.001). To test the functional contribution of OPN to lymphatic metastasis and to determine whether the RGD (Arg-Gly-Asp) integrin-binding sequence of OPN is important for this process, we transfected wild-type OPN or mutant OPN (lacking the RGD sequence) into MDA-MB-468 human breast cancer cells. In vitro, cells overexpressing OPN demonstrated increased anchorage-independent growth in soft agar (P = 0.001) and increased RGD-dependent adhesion (P = 0.045). Following mammary fat pad injection of nude mice, cells overexpressing OPN showed increased lymphovascular invasion, lymph node metastases, and lung micrometastases at earlier time points (P = 0.024). Loss of the RGD region partially abrogated this effect in the lymphatics (P = 0.038). These novel findings indicate that OPN is a key molecular player involved in lymphatic metastasis of breast cancer, potentially by affecting RGD-mediated adhesive interactions and by enhancing the establishment/persistence of tumor cells in the lymphatics.
View Article and Find Full Text PDF

Download full-text PDF

Source Listing
July 2006

TLD skin dose measurements and acute and late effects after lumpectomy and high-dose-rate brachytherapy only for early breast cancer.

Int J Radiat Oncol Biol Phys 2005 Aug;62(5):1283-90

Department of Radiation Oncology, London Regional Cancer Program, London, Ontario, Canada.

Purpose: This report examines the relationships between measured skin doses and the acute and late skin and soft tissue changes in a pilot study of lumpectomy and high-dose-rate brachytherapy only for breast cancer.

Methods And Materials: Thirty-seven of 39 women enrolled in this pilot study of high-dose-rate brachytherapy (37.2 Gy in 10 fractions b.i.d.) each had thermoluminescent dosimetry (TLD) at 5 points on the skin of the breast overlying the implant volume. Skin changes at TLD dose points and fibrosis at the lumpectomy site were documented every 6 to 12 months posttreatment using a standardized physician-rated cosmesis questionnaire. The relationships between TLD dose and acute skin reaction, pigmentation, or telangiectasia at 5 years were analyzed using the GEE algorithm and the GENMOD procedure in the SAS statistical package. Fisher's exact test was used to determine whether there were any significant associations between acute skin reaction and late pigmentation or telangiectasia or between the volumes encompassed by various isodoses and fibrosis or fat necrosis.

Results: The median TLD dose per fraction (185 dose points) multiplied by 10 was 9.2 Gy. In all 37 patients, acute skin reaction Grade 1 or higher was observed at 5.9% (6 of 102) of dose points receiving 10 Gy or less vs. 44.6% (37 of 83) of dose points receiving more than 10 Gy (p < 0.0001). In 25 patients at 60 months, 1.5% telangiectasia was seen at dose points receiving 10 Gy or less (1 of 69) vs. 18% (10 of 56) telangiectasia at dose points receiving more than 10 Gy (p = 0.004). Grade 1 or more pigmentation developed at 1.5% (1 of 69) of dose points receiving less than 10 Gy vs. 25% (14 of 56) of dose points receiving more than 10 Gy (p < 0.001). A Grade 1 or more acute skin reaction was also significantly associated with development of Grade 1 or more pigmentation or telangiectasia at 60 months. This association was most significant for acute reaction and telangiectasia directly over the lumpectomy site (p < 0.001). Grade 1 or more fibrosis, in 25 patients with a 60-month follow-up, occurred in 47.4% (9 of 19) of patients with a volume of 45 cm3 or less covered by the 100% isodose vs. 83.3% (5 of 6) of patients with a larger volume (p = 0.180). Asymptomatic and biopsy-proven fat necrosis occurred in 5 patients. No significant differences in fat necrosis rates according to volume were detected.

Conclusions: For high-dose-rate brachytherapy to the lumpectomy site, TLD skin dose was significantly related to acute skin reaction and to pigmentation and telangiectasia at 60 months. An acute skin reaction was also significantly associated with the development of telangiectasia at 60 months. TLD skin dose measurement may allow modification of the brachytherapy implant geometry (dwell times and position) to minimize late skin toxicity.
View Article and Find Full Text PDF

Download full-text PDF

Source Listing
August 2005

Patterns of breast recurrence in a pilot study of brachytherapy confined to the lumpectomy site for early breast cancer with six years' minimum follow-up.

Int J Radiat Oncol Biol Phys 2003 Dec;57(5):1239-46

Department of Radiation Oncology, London Regional Cancer Centre, London, Ontario, Canada.

Purpose: In this pilot study of high-dose-rate brachytherapy to the lumpectomy site as the sole radiation, ipsilateral and contralateral breast recurrences are documented with specific attention to the location of recurrence relative to the lumpectomy site.

Methods: Between March 1992 and January 1996, 39 patients with T1 (32 patients) and T2 breast cancers received 37.2 Gy in 10 fractions (b.i.d.) over 1 week prescribed to a volume encompassing the surgical clips. Thirteen received adjuvant tamoxifen, and 4 received chemotherapy. Follow-up included annual bilateral mammograms and clinical breast examination every 3 to 6 months. Whereas 13 patients had intraoperative implantation of the lumpectomy site, 26 had postoperative implantation. The latter group and 7 of the former group had surgical clips marking the lumpectomy site, which allowed estimates of the distance of any ipsilateral breast recurrence from the lumpectomy site, using the mediolateral and cranio-caudad mammographic views.

Results: At a median follow-up of 91 months, 33 women are alive, 4 have died of disease, and 2 have died of other causes. The 5-year actuarial rate of ipsilateral breast recurrence was 16.2%. Of 6 ipsilateral recurrences, 2 occurred within the lumpectomy site (in-field recurrences). One of the 2 patients had a 1-mm microscopic margin at initial diagnosis; the recurrence was a 3.5-mm microscopic focus of duct carcinoma in situ. The other patient had a 1.5-cm, high-grade infiltrating mammary carcinoma with no residual at wider resection at first diagnosis; the 5-mm invasive recurrence was also of high grade. Four women developed invasive recurrences at least 1.6 cm or more from the lumpectomy site (out-of-field recurrences). Two of these women had gross multifocal recurrences with two cancers in each patient; 1 of the 2 patients had an extensive intraductal component at initial diagnosis. The estimated nearest distances between the out-of-field recurrences and the surgical clips were 1.6, 5.5, 7.7, and 12.0 cm. All ipsilateral breast recurrences were salvaged by mastectomy (4 patients) or by repeat lumpectomy (2 patients) and whole-breast radiation. The interval postdiagnosis to ipsilateral recurrence ranged from 20 months to 58 months. There were two contralateral breast recurrences at intervals of 34 and 36 months; 1 of these patients also had a multifocal, ipsilateral recurrence at 58 months, as previously described. Among patients with any breast recurrence, 1 patient had a family history of prostate cancer; there was no family history of breast or ovarian cancer. Of 17 patients who received adjuvant systemic therapy, only 1 had a breast recurrence.

Conclusions: In this pilot study, breast recurrences outside of the lumpectomy site were the predominant pattern of recurrence.
View Article and Find Full Text PDF

Download full-text PDF

Source Listing
December 2003