arXiv subjects

Dennis Della Corte

Publications and source records attributed to Dennis Della Corte.

4 recordsLinked to original sources

arXiv2025

PANDA-PLUS-Bench: A Clinical Benchmark for Evaluating Robustness of AI Foundation Models in Prostate Cancer Diagnosis

Artificial intelligence foundation models are increasingly deployed for prostate cancer Gleason grading, where GP3/GP4 distinction directly impacts treatment decisions. However, these models may achieve high validation accuracy by learning specimen-specific artifacts rather than generalizable biological features, limiting real-world clinical utility. We introduce PANDA-PLUS-Bench, a curated benchmark dataset derived from expert-annotated prostate biopsies designed specifically to quantify this failure mode. The benchmark comprises nine carefully selected whole slide images from nine unique patients containing diverse Gleason patterns, with non-overlapping tissue patches extracted at both 512x512 and 224x224 pixel resolutions across eight augmentation conditions. Using this benchmark, we evaluate seven foundation models on their ability to separate biological signal from slide-level confounders. Our results reveal substantial variation in robustness across models: Virchow2 achieved the lowest slide-level encoding among large-scale models (81.0%) yet exhibited the second-lowest cross-slide accuracy (47.2%). HistoEncoder, trained specifically on prostate tissue, demonstrated the highest cross-slide accuracy (59.7%) and the strongest slide-level encoding (90.3%), suggesting tissue-specific training may enhance both biological feature capture and slide-specific signatures. All models exhibited measurable within-slide vs. cross-slide accuracy gaps, though the magnitude varied from 19.9 percentage points to 26.9 percentage points. We provide an open-source Google Colab notebook enabling researchers to evaluate additional foundation models against our benchmark using standardized metrics. PANDA-PLUS-Bench addresses a critical gap in foundation model evaluation by providing a purpose-built resource for robustness assessment in the clinically important context of Gleason grading.

cs.CV↗

arXiv2024

Overcoming Limitations in Artificial Intelligence-based Prostate Cancer Detection through Better Datasets and a Bayesian Approach to Aggregate Panel Predictions

Despite considerable progress in developing artificial intelligence (AI) algorithms for prostate cancer detection from whole slide images, the clinical applicability of these models remains limited due to variability in pathological annotations and existing dataset limitations. This article proposes a novel approach to overcome these challenges by leveraging a Bayesian framework to seamlessly integrate new data, and present results as a panel of annotations. The framework is demonstrated by integrating a Bayesian prior with one trained AI model to generate a distribution of Gleason patterns for each pixel of an image. It is shown that using this distribution of Gleason patterns rather than a ground-truth label can improve model applicability, mitigate errors, and highlight areas of interest for pathologists. Additionally, we present a high-quality, hand-curated dataset of prostate histopathological images annotated at the gland level by trained pre-medical students and verified by an expert pathologist. We highlight the potential of this adaptive and uncertainty-aware framework for developing clinically deployable AI tools that can support pathologists in accurate prostate cancer grading, improve diagnostic accuracy, and create positive patient outcomes.

q-bio.TO↗

arXiv2022

Learning Small Molecule Energies and Interatomic Forces with an Equivariant Transformer on the ANI-1x Dataset

Accurate predictions of interatomic energies and forces are essential for high quality molecular dynamic simulations (MD). Machine learning algorithms can be used to overcome limitations of classical MD by predicting ab initio quality energies and forces. SE(3)-equivariant neural network allow reasoning over spatial relationships and exploiting the rotational and translational symmetries. One such algorithm is the SE(3)-Transformer, which we adapt for the ANI-1x dataset. Our early experimental results indicate through ablation studies that deeper networks - with additional SE(3)-Transformer layers - could reach necessary accuracies to allow effective integration with MD. However, faster implementations of the SE(3)-Transformer will be required, such as the recently published accelerated version by Milesi.

physics.chem-ph↗

arXiv2020

Student created video content for assessment, experiential learning, and service

This paper outlines the feasibility of replacing a midterm assessment with a student-created digital media project. We examine the benefits of a peer review process for student-created pedagogy, the effectiveness of the student-generated digital media, and self-evaluations as a replacement for accepted forms of assessment in terms of: learning, assessment, and service provided. The outlined project also facilitates the creation of a video bank easily accessible for professors and students alike in introductory physics courses using the OpenStax College Physics textbook.1 We conclude that this non-traditional teaching and assessment model is both effective and timely as more college courses are being taught digitally and/or asynchronously.

physics.ed-ph↗