Hot off the press 🔥 Fred Howard and colleagues published a research article in #NPJBreastCancer demonstrating the utility of #DeepLearningTools in predicting the risk of recurrence in #BreastCancer patients. UChicagoCancerCenter UChicago Ob/Gyn ow.ly/pAA950NJlQJ

Validating a low-cost, open-source, locally manufactured workstation and computational pipeline for automated histopathology evaluation using deep learning biorxiv.org/cgi/content/sh… #bioRxiv

In an analysis of over one million pts with #breastcancer, HER2-low status had minimal impact on breast cancer survival & chemotherapy response when compared to HER2-0 disease in an analysis prior to introduction of potent antibody-drug conjugates. ja.ma/3KPOtzT #BCSM

Now out in npj Digital Medicine! We found that #ChatGPT could generate scientific abstracts that were often difficult to distinguish from real abstracts. Hallucinates though, so needs careful review! Cheers & thanks: Fred Howard Nikolay Markov @[email protected] Emma Dyer Sid Yuan Luo Pearson Lab

Topical content... Blinded human reviewers only detected 68% of #ChatGPT generated scientific abstracts, despite incorrectly identifying 14% of real abstracts as being generated. nature.com/articles/s4174…

Overjoyed to see CarpeDiem out in Journal of Clinical Investigation! We used #machinelearning to examine unresolving pneumonia here: jci.org/articles/view/… NM Pulmonary and Thoracic Surgery Northwestern Feinberg School of Medicine NIAID News Nikolay Markov @[email protected] macwalter Rogan Grant @[email protected] Yuan Luo @ScottBudinger Alexander Misharin Benjamin D. Singer

Our latest paper, "Deep learning generates synthetic cancer histology for explainability and education", is out! nature.com/articles/s4169…. We use #digitalpathology AI for 🧪 scientific discovery & 🏫 education by enabling deep neural networks to "teach back" what they learn. (1/5)

Augustin Vannier discusses the data in today’s poster session at #ASCO23. He explains the utilization of genomic data to identify the patients that benefit from chemotherapy treatment. Fred Howard UChicagoCancerCenter UChicago Medicine University of Chicago Hematology/Oncology

Absolutely outstanding discussion of AI based predictors specifically around lobular breast cancer by Fred Howard UChicago Medicine following amazing work by Fresia Pareja in the Reis Filho lab

Excited to share our study of RSClin in 280k pts doi.org/10.1002/cncr.3… In the lowest risk pts per RSClin-no chemo benefit even if Oncotype high. For high risk-question of OFS vs chemo remains (I vote OFS) Congrats Augustin! Genomic Health® Olufunmilayo Olopade Rita Nanda UChicago Medicine Breast Cancer Program

UChicago Pritzker School of Medicine Fred Howard ASCO Liz Garrett-Mayer, PhD OpenAI @jcocci_asco will publish Fred Howard Liz Garrett-Mayer, PhD work "Characterizing the Increase in Artificial Intelligence Content Detection in Oncology Scientific Abstracts from 2021 to 2023" and #JCOCCI EIC Dr. Jeremy Warner to discuss live. 📍June 2, 10:00AM S403B. (Journals Lounge) (4/n)

‼️ #ASCO24 simultaneous publication by Pearson Lab et al: Characterizing the Increase in Artificial Intelligence Content Detection in Oncology Scientific Abstracts from 2021 to 2023: ➡️brnw.ch/21wKljc

Fred Howard presents an excellent poster at #ASCO24 showing increased use of ChatGPT in generating scientific abstracts directed towards need to bring new policies. @Lab_Pearson UChicagoCancerCenter University of Chicago Hematology/Oncology

Exciting work in npj Journals from UChicago Pritzker School of Medicine's Asim Dhungana - predicting recurrence from quantitative IHC using #machinelearning, prognostic in Oncotype high/low risk cases, calculator at rsncdb.cri.uchicago.edu UChicagoCancerCenter UChicago Medicine Breast Cancer Program #bcsm doi.org/10.1038/s41523…