Dr. Margaret Brandwein-Weber Profile

Dr. Margaret Brandwein-Weber

at Icahn School of Medicine at Mount Sinai

SPIE Involvement:

Author

Publications (8)

Proceedings Article | 3 April 2024 Presentation + Paper

Federated learning in computational pathology: classification of tall cell patterns in papillary thyroid carcinoma

Sonal Shukla, Margaret Brandwein-Weber, Shabnam Samankan, Ahmed Ayad, Mohamed Rabie, Scott Doyle

Proceedings Volume 12933, 129330U (2024) https://doi.org/10.1117/12.3006890

KEYWORDS: Data modeling, Tumor growth modeling, Pathology, Data privacy, Artificial intelligence, Thyroid, Deep learning, Performance modeling, Computer vision technology, Machine learning, Image classification, Computer aided detection

Read Abstract +

This paper presents an exploration of Federated Learning (FL) in medical imaging, focusing on Computational Pathology (CP) with Whole Slide Images (WSIs) for head and neck cancer. While previous FL approaches in healthcare targeted radiology, genetics, and Electronic Health Records (EHRs), our research addresses the understudied area of CP datasets. Our aim is to develop robust AI models for CP datasets without sacrificing data privacy and security. To this end, we demonstrate the use of FL applied to a CP dataset consisting of papillary thyroid carcinoma, specifically focusing on the rare and aggressive variant called Tall Cell Morphology (TCM). Patients with TCM require more aggressive treatment and rigorous follow-up due to its aggressiveness and increased recurrence rates. In this work, we perform a simulated FL training experiment by dividing a dataset into three virtual ”clients”. We locally train a Convolutional Neural Network (CNN), to classify patches of tissue labelled from the local WSI dataset as “tall” (expressing TCM) or “non-tall”. Models are then aggregated and convergence is ensured through the Federated Averaging (FedAVG) algorithm. The decentralized approach of FL creates a secure and privacy-preserving collaborative training environment, keeping individual client data local through horizontal data partitioning. This enables collective training of deep learning models on distributed data, benefiting from a diverse and rich dataset while safeguarding patient privacy. We compare the efficacy of the FL-trained model to a centralized model (trained using all ”client” data together) using accuracy, sensitivity, specificity, and F-1 score. Our findings indicate that the simulated FL models exhibit performance on par with or superior to centralized learning, achieving accuracy scores between 75-87%, while centralized learning attains an accuracy of 82%. This novel approach holds promise for revolutionizing computational pathology and contributing to more effective medical decision-making.

Proceedings Article | 7 April 2023 Presentation + Paper

Combining multiple ground truth annotations for segmentation training for oral cavity cancer

Jonathan Folmsbee, Margaret Brandwein-Weber, Scott Doyle

Proceedings Volume 12471, 1247112 (2023) https://doi.org/10.1117/12.2654301

KEYWORDS: Tumors, Image segmentation, Education and training, Tissues, Image fusion, Cancer, Semantics, Pathology, Artificial intelligence, Data modeling

Read Abstract +

Proceedings Article | 4 April 2022 Presentation + Paper

Automatic flagging of AI segmentation errors in computational pathology

Rakesh Choudhary, Dhadma Balachandran, Jonathan Folmsbee, Jawaria Rahman, Margaret Brandwein, Scott Doyle

Proceedings Volume 12039, 1203907 (2022) https://doi.org/10.1117/12.2613194

KEYWORDS: Image segmentation, Tumors, Artificial intelligence, Satellites, Tissues, Feature extraction, Pathology, Evolutionary algorithms, Statistical analysis, Cancer

Read Abstract +

Proceedings Article | 4 April 2022 Presentation

Cloud-based platform for human-in-the-loop re-annotation of whole slide imaging: large scale semantic pathology segmentation

Jonathan Folmsbee, Scott Doyle, Rakesh Choudhary, Margaret Brandwein-Weber, Jawaria Rahman

Proceedings Volume 12039, 1203912 (2022) https://doi.org/10.1117/12.2613253

KEYWORDS: Image segmentation, Pathology, Tumors, Image resolution, Digital imaging

Read Abstract +

Proceedings Article | 15 February 2021 Poster + Presentation + Paper

Comparing architectural features between heuristically human-annotated and Artificial Intelligence (AI) generated tumor and satellite labels in early-stage oral cavity cancer

Dhadma Balachandran, Margaret Brandwein-Weber, Jonathan Folmsbee, Scott Doyle

Proceedings Volume 11603, 116030X (2021) https://doi.org/10.1117/12.2581153

KEYWORDS: Tumors, Satellites, Cancer, Machine vision, Human vision and color perception, Computer vision technology, Tissues

Read Abstract +

Proceedings Article | 15 February 2021 Presentation + Paper

Whole slide semantic segmentation: large scale active learning for digital pathology

Jonathan Folmsbee, Margaret Brandwein-Weber, Scott Doyle

Proceedings Volume 11603, 116030G (2021) https://doi.org/10.1117/12.2581229

KEYWORDS: Pathology, Image segmentation, Data modeling

Read Abstract +

Proceedings Article | 18 March 2019 Paper

Fragile neural networks: the importance of image standardization for deep learning in digital pathology

Jonathan Folmsbee, Starr Johnson, Xulei Liu, Margaret Brandwein-Weber, Scott Doyle

Proceedings Volume 10956, 1095613 (2019) https://doi.org/10.1117/12.2512992

KEYWORDS: Tissues, Pathology, Image processing, 3D modeling, Neural networks, Image classification, Tumors, RGB color model, Digital imaging, Light sources and illumination

Read Abstract +

Proceedings Article | 9 March 2018 Paper

Registration parameter optimization for 3D tissue modeling from resected tumors cut into serial H and E slides

Starr Johnson, Margaret Brandwein, Scott Doyle

Proceedings Volume 10581, 105810T (2018) https://doi.org/10.1117/12.2293962

KEYWORDS: Image registration, 3D modeling, Tissues, Tumors, Image filtering, Optimization (mathematics), Image compression, Cancer, Pathology, Surgery

Read Abstract +

Showing 5 of 8 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years