# Comprehensive framework for evaluation of deep neural networks in detection and quantification of lymphoma from PET/CT images: clinical insights, pitfalls, and observer agreement analyses

arXiv: https://arxiv.org/abs/2311.09614 · HTML: https://mlsanity.com/paper/2311.09614/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 55 references that are in the map.
- First posted on arXiv 16 Nov 2023.

## Builds on (most cited first, 11 of 11)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images](https://mlsanity.com/paper/2201.01266/) (2022)
- [3D MRI brain tumor segmentation using autoencoder regularization](https://mlsanity.com/paper/1810.11654/) (2018)
- [nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation](https://mlsanity.com/paper/1809.10486/) (2018)
- [A Survey on Active Learning and Human-in-the-Loop Deep Learning for Medical Image Analysis](https://mlsanity.com/paper/1910.02923/) (2019)
- [Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening](https://mlsanity.com/paper/1903.08297/) (2019)
- [Multimodal Spatial Attention Module for Targeting Multimodal PET-CT Lung Tumor Segmentation](https://mlsanity.com/paper/2007.14728/) (2020)
- [D-LEMA: Deep Learning Ensembles from Multiple Annotations -- Application to Skin Lesion Segmentation](https://mlsanity.com/paper/2012.07206/) (2020)
- [Mirror U-Net: Marrying Multimodal Fission with Multi-task Learning for Semantic Segmentation in Medical Imaging](https://mlsanity.com/paper/2303.07126/) (2023)
- [Automated head and neck tumor segmentation from 3D PET/CT](https://mlsanity.com/paper/2209.10809/) (2022)
- [MCTNet: A Multi-Scale CNN-Transformer Network for Change Detection in Optical Remote Sensing Images](https://mlsanity.com/paper/2210.07601/) (2022)

## Built on by (most cited first, 4 of 4)

- [AutoPET Challenge III: Testing the Robustness of Generalized Dice Focal Loss trained 3D Residual UNet for FDG and PSMA Lesion Segmentation from Whole-Body PET/CT Images](https://mlsanity.com/paper/2409.10151/) (2024)
- [Adaptive Voxel-Weighted Loss Using L1 Norms in Deep Neural Networks for Detection and Segmentation of Prostate Cancer Lesions in PET/CT Images](https://mlsanity.com/paper/2502.02756/) (2025)
- [IgCONDA-PET: Weakly-Supervised PET Anomaly Detection using Implicitly-Guided Attention-Conditional Counterfactual Diffusion Modeling -- a Multi-Center, Multi-Cancer, and Multi-Tracer Study](https://mlsanity.com/paper/2405.00239/) (2024)
- [Comprehensive Evaluation of Quantitative Measurements from Automated Deep Segmentations of PSMA PET/CT Images](https://mlsanity.com/paper/2504.16237/) (2025)

## Abstract (arXiv)

This study addresses critical gaps in automated lymphoma segmentation from PET/CT images, focusing on issues often overlooked in existing literature. While deep learning has been applied for lymphoma lesion segmentation, few studies incorporate out-of-distribution testing, raising concerns about model generalizability across diverse imaging conditions and patient populations. We highlight the need to compare model performance with expert human annotators, including intra- and inter-observer variability, to understand task difficulty better. Most approaches focus on overall segmentation accuracy but overlook lesion-specific measures important for precise lesion detection and disease quantification. To address these gaps, we propose a clinically relevant framework for evaluating deep segmentation networks. Using this lesion measure-specific evaluation, we assess the performance of four deep networks (ResUNet, SegResNet, DynUNet, and SwinUNETR) across 611 cases from multi-institutional datasets, covering various lymphoma subtypes and lesion characteristics. Beyond standard metrics like the Dice similarity coefficient, we evaluate clinical lesion measures and their prediction errors. We also introduce detection criteria for lesion localization and propose a new detection Criterion 3 based on metabolic characteristics. We show that networks perform better on large, intense lesions with higher metabolic activity. Finally, we compare network performance to physicians via intra- and inter-observer variability analyses, demonstrating that network errors closely resemble those made by experts, i.e., the small and faint lesions remain challenging for both humans and networks. This study aims to improve automated lesion segmentation's clinical relevance, supporting better treatment decisions for lymphoma patients. The code is available at: https://github.com/microsoft/lymphoma-segmentation-dnn.