MoA: Mixture of Aggregators Improves Slide-Level Diagnosis in Computational Pathology

Fatih Ibrahim Ozlugedik, Muhammed Furkan Dasdelen, Rao Muhammad Umer, Carsten Marr

Keywords: computational pathology, multiple instance learning, cytology

TL;DR: MoA: Mixture of Aggregators

Abstract: Multiple instance learning (MIL) is the standard for learning slide-level representations from whole slide images (WSIs), typically using a single attention-based aggregator to pool instance features. However, a single aggregator can struggle to capture the diverse morphological patterns in heterogeneous pathology and cytology data, where different diseases may demand different pooling behaviours. We propose a mixture-of-aggregators framework that models complementary aspects of instance distributions in histology and hematologic cytology. A router with top-2 gating dynamically selects the most relevant aggregators per slide, and their outputs are fused into a patient-level representation. To avoid collapse to a single dominant expert, we add a load-balancing loss and Gumbel noise on the router logits to promote the use of multiple aggregators. We extensively evaluate our method on 19 different tasks derived from 16 datasets including histology and hematologic cytology. Compared to single-aggregator baselines, our approach improves diagnostic prediction accuracy by an average of 4.5\% over ABMIL and 12.6\% over TransMIL across all tasks. Beyond performance, our analysis shows that different aggregators attend to distinct, disease-specific instance distributions, providing interpretable insights into the diagnostic process.

Primary Subject Area: Application: Histopathology

Secondary Subject Area: Detection and Diagnosis

Registration Requirement: Yes

Reproducibility: https://github.com/fatihOzlugedik/MixtureOfAggregators/tree/main

Visa & Travel: No

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 238