Machine learning improves analysis of weakly labeled bioimages for ALS studies
Model-agnostic Knowledge Augmented Machine Learning for Weakly Annotated Bioimaging Data
Artificial IntelligenceComputer Vision and Pattern Recognition
Summary
Machine learning models often struggle to analyze biological images that lack detailed labels, like those showing cell structures. The authors present a new method that adds expert knowledge to these models, helping them better recognize important features. This approach improved the accuracy of models analyzing cell images linked to ALS, a neurodegenerative disease. It also revealed biological effects of stressors and drugs on a protein related to ALS, potentially advancing disease understanding and treatment research.
What this means in practice
- •For bioinformatics software developers: Improve image analysis tools for cellular pathology by integrating domain knowledge to enhance weakly annotated bioimaging classification accuracy.
- •For pharmaceutical research teams: Use enhanced machine learning pipelines to quantify phenotypes linked to neurodegenerative diseases for drug effect assessment in cellular microscopy images.
Authors
Jayanie Bogahawatte, Sachith Seneviratne, Simon Maksour, Dzung Do-Ha, Elizabeth Qian, Samuel Brennan, Matthew Keon, Lezanne Ooi, Saman Halgamuge
Abstract
Machine learning models are increasingly used for automated feature extraction from weakly annotated bioimaging data, such as microscopy and cytomorphology images. Due to the sole dependency of automated feature extraction, existing methods yield inconsistent classification performance across weakly annotated datasets. Here, we propose a model-agnostic knowledge augmentation method that enables ML models to exploit domain knowledge beyond what automated feature extraction can capture. The proposed approach improves classification performance across seven ML architectures. When integrated with masked autoencoder-based self-supervised learning, our method achieves the strongest overall performance, improving precision by 2.9% over the next-best method on three in-house microscopy datasets and by an average of 6.8% across three public cytomorphology benchmarks. We perform nucleus-level classification based on mislocalization of TAR DNA-binding protein 43 (TDP-43) from the nucleus to the cytoplasm, a phenotype linked to Amyotrophic Lateral Sclerosis (ALS), using three independent microscopy datasets of motor neurons derived from human induced pluripotent stem cells. Accurate quantification of this phenotype at scale remains challenging and may contribute to inconsistencies across studies. Application of the best-performing model to these microscopy datasets reveals biological interpretations related to ALS, including dose-dependent effects of the proteasomal stressor, MG132, and the oxidative stressor, sodium arsenite (SA), on TDP-43 mislocalization, variable pathology in motor neurons from sporadic and familial ALS donors, and a lack of protection of the pharmacological compounds, edaravone and niclosamide, against SA-induced TDP-43 mislocalization. Together, our pipeline enables in-vitro cellular image analysis to advance understanding of ALS-associated pathology and potential treatments.