MILCA: Malaria Parasite Detection from Sample-Level Weak Labels

Petru Manescu, Delmiro Fernandez-Reyes

Keywords: malaria, weakly supervised object detection, microscopy, multiple instance learning, class activations

TL;DR: Weakly supervised malaria parasite detection with sample level labels.

Abstract: Malaria diagnosis requires the inspection of multiple image fields per sample. Training vision models for malaria parasite detection typically requires large numbers of expert-provided bounding boxes, which are costly to obtain and often impractical in real-world deployments. We introduce MILCA, a weakly supervised object detection framework that learns parasite localization from sample-level diagnostic labels, which are routinely recorded in clinical practice. MILCA combines Multiple Instance Learning (MIL) for sample classification with an iterative Class Activation (CA) Mapping procedure that yields coarse parasite pseudo-labels, which are further enriched with hard negatives from parasite-free samples. These pseudo-labels enable training a detector without any manual bounding-box supervision. Experiments on multiple microscopy datasets show that MILCA achieves reliable detection and counting performance under fully weak supervision, and that fine-tuning with only a small fraction of expert annotations provides substantial additional gains, outperforming supervised and pseudo-labeling baselines under the same or lower annotation budgets. By converting coarse, sample-level clinical labels into effective object-level supervision and leveraging a hybrid refinement scheme, MILCA provides a label-efficient route toward automated malaria parasite and detection and a general approach for weakly supervised blood film analysis.

Primary Subject Area: Detection and Diagnosis

Secondary Subject Area: Learning with Noisy Labels and Limited Data

Registration Requirement: Yes

Reproducibility: www.github/UCL/MILCA

Visa & Travel: Yes

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 127