Przejdź do treści Wyszukiwarka
dla student贸w, dla biznesu, dla instytucji naukowych
15.11.2023 2:00 pm - 15.11.2023 3:00 pm

Abstrakt:

Prognostic and Predictive uses of an AI Morphometric Digital Pathology Platform

The application of evidence-based precision medicine to cancer patients results in the most efficacious results with lower toxicity. However, cancer is a heterogeneous disease and different patients respond differently to the same treatment both in terms of outcomes and side effects. Current predictive and prognostic biomarkers have limited accuracy.

The application of Artificial Intelligence offers the potential of significantly developing new biomarkers with improved accuracy. Standard H&E pathology slides from biopsies and surgical specimens have been used for many years. Each slide contains a wealth of detailed information well beyond the ability of the human eye to synthesize and comprehend. AI offers the opportunity to unveil the tremendous information embedded in each slide. A proprietary, cloud based, AI platform (PathomIQ) will be presented which enables accurate diagnosis and grading, as well as, the capability to predict response to specific treatments in major cancers such as prostate cancer and triple negative breast cancer. Prognostically, the PathomIQ platform can predict progression to biochemical relapse and development of metastases after prostatectomy with greater than 80% accuracy, which is significantly better than the accuracy of current genomic panels and algorithms in current use. It can also predict who will most likely benefit from the addition of antiandrogen to treatment regimens and who with TNBC will likely achieve a pathologic complete response from neoadjuvant therapy. 

Additionally, molecular interrogation of regions of interest identified by AI can be used to discover new molecular biomarkers and the AI platform can predict the presence of common genomic changes with high accuracy from H&E slides. The power of the AI platform can assist physicians in selecting optimal therapy and improve the efficiency in patient selection for clinical trials.

Biogram:

George Wilding, MD graduated with an MD from the University of Massachusetts Medical School where he also trained in Internal Medicine. Previously, he had received a BS in Chemical Engineering, a BS in Biology from Tufts University and an MS in Pharmacology from Penn State University. He received training in Medical Oncology in the Medicine Branch at the NCI. He served on the faculty of the University of Wisconsin-Madison from 1988 to 2015. During this period he progressed from Assistant Professor to Professor of Medicine.

His research focused in the laboratory and clinic on prostate carcinogenesis and the development of new therapies for advanced prostate cancer. He has had continuous federal peer reviewed funding for approximately 30 plus years. During his time at the University of Wisconsin, he served as the Head of Hematology/Oncology for 18 years and as Director of the Carbone Cancer Center, an NCI designated comprehensive cancer center, for approximately 12 years. During his time as Director of the UWCCC he built strong relations and collaborations with academic cancer centers in China, Taiwan and South Korea. While at the University of Texas MD Anderson Cancer Centerrom September 2015 to January 2021, Dr. Wilding served as Vice President and Vice Provost/Deputy Chief Academic Officer of Clinical and Interdisciplinary Research, Professor of Genitourinary Medical Oncology and Associate Director for Interdisciplinary Research on the NCI Cancer Center Support Grant for MD Anderson.

He was also part of the Moon Shot Leadership Team and was active in the Global Academic Program with activities in Poland, Hungary, China and Mexico. Currently, he lives in Albuquerque, NM and is an active advisor/consultant for several biotech companies, the NCI Clinical Trial Advisory Committee and MD Anderson and is a member of the Polish Medical Research Agency Advisory Council.


Zach臋camy do udzia艂u w wydarzeniu stacjonarnie 鈥 prosimy o wiadomo艣膰 mailow膮 na adres registration@ideas-ncbr.pl z podaniem imienia, nazwiska i afiliacji. Mo偶liwy jest te偶 udzia艂 przez platform臋 Zoom:

Link Zoom: https://us06web.zoom.us/j/8091539654?pwd=K3NvVDZZUVpKYVRMS1VWU1poWW1EZz09

ID: 809 153 9654

Passcode: 712966

20.09.2023
Webinaria dla kandydat贸w do szk贸艂 doktorskich i promotor贸w
20.05.2023
OpenAI Meeting
01.06.2023
Neural Fields in聽Computer Graphics. Wyk艂ad Przemys艂awa Musialskiego