Detection versus Instance Segmentation for Multi-Species Malaria Diagnosis: A Head-to-Head Comparison and Multi-Dataset Validation of YOLOv12 Architectures with Small Object Optimization

Ahmed Tahiru Issah, Idaya Seidu, Carine Mukamakuza

Keywords: malaria detection, instance segmentation, object detection, YOLOv12, thick blood smear, validation study, head-to-head comparison, Plasmodium falciparum, small object detection, deep learning

TL;DR: Head-to-head validation comparing YOLOv12 detection and segmentation variants with P2 small-object heads for multi-species malaria diagnosis, with per-species fairness analysis.

Abstract: Automated malaria parasite detection using deep learning holds promise for addressing diagnostic gaps in resource-limited settings, yet most studies rely on single-dataset evaluations that fail to capture real-world variability. In this work, we rigorously validate YOLOv12-based architectures for malaria detection across diverse geographic and institutional contexts. We introduce a dual-head architecture combining instance segmentation with a high-resolution P2 detection head to target tiny ring-stage parasites. Our evaluation on a diverse Rwandan thick-smear dataset (2,739 images) and two external datasets from Ghana (Lacuna) and Nigeria (FASTMAL) reveals critical insights into model robustness. While the proposed YOLOv12-Seg-N-P2 model achieves state-of-the-art internal performance (mAP@50 0.888) and significantly improves detection of challenging \textit{P. vivax} (+10.9\%) and \textit{P. falciparum} ring forms, external validation exposes severe domain shift, with performance dropping by $>$80\% on unseen datasets. We further demonstrate that while P2 heads enhance morphological precision on source data, they reduce zero-shot generalization, likely by overfitting to dataset-specific acquisition characteristics. Our findings underscore the necessity of multi-center validation and suggest that deployment-ready malaria AI requires domain adaptation strategies rather than purely architectural innovations.

Primary Subject Area: Detection and Diagnosis

Secondary Subject Area: Segmentation

Registration Requirement: Yes

Visa & Travel: Yes

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 42