# Self-training of Machine Learning Models for Liver Histopathology: Generalization under Clinical Shifts

arXiv: https://arxiv.org/abs/2211.07692 · HTML: https://mlsanity.com/paper/2211.07692/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 21 references that are in the map.
- First posted on arXiv 14 Nov 2022.

## Builds on (most cited first, 7 of 7)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks](https://mlsanity.com/paper/1905.11001/) (2019)
- [In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learning](https://mlsanity.com/paper/2101.06329/) (2021)
- [Deep neural network models for computational histopathology: A survey](https://mlsanity.com/paper/1912.12378/) (2019)
- [Teacher-Student chain for efficient semi-supervised histology image classification](https://mlsanity.com/paper/2003.08797/) (2020)
- [Rethinking Machine Learning Model Evaluation in Pathology](https://mlsanity.com/paper/2204.05205/) (2022)

## Built on by (most cited first, 2 of 2)

- [PLUTO: Pathology-Universal Transformer](https://mlsanity.com/paper/2405.07905/) (2024)
- [SC-MIL: Supervised Contrastive Multiple Instance Learning for Imbalanced Classification in Pathology](https://mlsanity.com/paper/2303.13405/) (2023)

## Abstract (arXiv)

Histopathology images are gigapixel-sized and include features and information at different resolutions. Collecting annotations in histopathology requires highly specialized pathologists, making it expensive and time-consuming. Self-training can alleviate annotation constraints by learning from both labeled and unlabeled data, reducing the amount of annotations required from pathologists. We study the design of teacher-student self-training systems for Non-alcoholic Steatohepatitis (NASH) using clinical histopathology datasets with limited annotations. We evaluate the models on in-distribution and out-of-distribution test data under clinical data shifts. We demonstrate that through self-training, the best student model statistically outperforms the teacher with a 3% absolute difference on the macro F1 score. The best student model also approaches the performance of a fully supervised model trained with twice as many annotations.