NSF MRI: High Performance Digital Pathology Using Big Data and Machine Learning

What's New?

  • (20220113) We are pleased to announce v2.0.0 of the TUH DPATH Corpus. This release contains a breast tissue subset and is described here. There are 3,505 annotated images partitioned into training, development and blind evaluation sets.

  • (20211215) We have completed scanning the Fox Chase Cancer Center subset and now have scanned over 80,000 slides.

  • (20200115) We reached an agreement with Fox Chase Cancer Center to scan all the sldies in their Biosample Repository Facility.

  • (20200115) We continue to scan slides for the TUH Tumor Board sessions where revered pathologists discuss cancer cases with our high definition digital slides.

  • (20191208) We presented our progess of our 1M digital slide database and ran live demos of our annotations at IEEE SPMB 2019. View our conference publication here.

  • (20191120) Under the mentorship of Dr. Jhala, Director of Anatomic Pathology and Cytology at TUH, we are creating a pilot corpus of breast cancer samples. We annotate four different features: ductal carcinoma in situ (DCIS), invasive ductal carcinoma (IDC), benign tissue, and inflammation.

  • (20191113) We are pleased to announce that we have reached a new milestone of 65,000 scanned slides, and are recognizing fifteen tissue types.

  • (20181228) Annotation Tutorial: A short video that demonstrates how we annotate images using Aperio's ImageScope software.

  • (20180727) The digital pathology team had a training session with the Aperio personnel on-site. The eSlideManager has been configured to handle our workflow, and now our system is fully functional.

  • (20180523) A sample of our digital pathology database, v0.0.0 is now available. This sample contains 6 patients and 3 tumor types (breast, gastrointestinal, urology).

  • (20180521) The app server that hosts Leica/Aperio's eSlide Manager software has been successfully installed.

  • (20180515) We passed a major milestone by completing scanning of over 3,000 pathology slides. These digital slides will be crucial in training our deep-learning system, as well as creating our database. Learn more here.

  • (20180208) Our digital scanner is now operational. This scanner will help create an open-source database of pathology slides. Learn more about our MRI project here.

  • (20180123) Our Leica Aperio AT2 has arrived at our facility. This is our digital scanner that we will be using to scan pathology slides. To learn more about our scanner, click here.