Beyond attention heatmaps: How to get better explanations for multiple instance learning models in histopathology.

Mina Jamshidi Idaji¹, Julius Hense², Tom Neuhäuser², Augustin Krause³, Yanqing Luo³, Oliver Eberle², Thomas Schnake², Laure Ciernik², Farnoush Rezaei Jafari², Reza Vahidimajd⁴, Jonas Dippel⁵, Christoph Walz⁶, Frederick Klauschen⁷, Andreas Mock⁸, Klaus-Robert Müller⁹

¹Berlin Institute for the Foundations of Learning and Data, Berlin, Germany; Machine Learning Group, Technische Universität Berlin, Berlin, Germany. Electronic address: mina.jamshidi.idaji@tu-berlin.de.

²Berlin Institute for the Foundations of Learning and Data, Berlin, Germany; Machine Learning Group, Technische Universität Berlin, Berlin, Germany.

³Machine Learning Group, Technische Universität Berlin, Berlin, Germany.

⁴Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong.

⁵Berlin Institute for the Foundations of Learning and Data, Berlin, Germany; Machine Learning Group, Technische Universität Berlin, Berlin, Germany; Aignostics, Berlin, Germany.

⁶Institute of Pathology, Ludwig Maximilian University, Munich, Germany.

⁷Institute of Pathology, Ludwig Maximilian University, Munich, Germany; German Cancer Research Center, Heidelberg, and German Cancer Consortium, Munich, Germany; Institute of Pathology, Charité Universitätsmedizin, Berlin, Germany.

⁸Institute of Pathology, Ludwig Maximilian University, Munich, Germany; German Cancer Research Center, Heidelberg, and German Cancer Consortium, Munich, Germany.

⁹Berlin Institute for the Foundations of Learning and Data, Berlin, Germany; Machine Learning Group, Technische Universität Berlin, Berlin, Germany; Department of Artificial Intelligence, Korea University, Seoul, Republic of Korea; Max-Planck Institute for Informatics, Saarbrücken, Germany.

PubMed DOI

Abstract

Multiple instance learning (MIL) has enabled substantial progress in computational histopathology, where a large amount of patches from gigapixel whole slide images are aggregated into slide-level predictions. Heatmaps are widely used to validate MIL models and to discover tissue biomarkers. Yet, the validity of these heatmaps has barely been investigated. In this work, we introduce a general fram…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free Account Sign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.