A dataset of medication images with instance segmentation masks for preventing adverse drug events

2026-03-11

AI Summary

This study addresses medication errors in real-world settings caused by the scarcity of annotated tablet image data. To this end, the authors introduce MEDISEG, a novel dataset featuring instance-level segmentation annotations for 8,262 images across 32 tablet classes, explicitly capturing challenging conditions such as occlusion, overlap, and illumination variations. Leveraging this dataset, the authors employ YOLOv8 and YOLOv9 models under both fully supervised training and few-shot transfer learning paradigms, achieving mAP@0.5 scores of 99.5% and 80.1% on the 3-Pills and 32-Pills subsets, respectively. Experimental results demonstrate that MEDISEG substantially enhances model generalization in complex multi-tablet scenarios and improves recognition performance on previously unseen tablet categories.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Medication errors and adverse drug events (ADEs) pose significant risks to patient safety, often arising from difficulties in reliably identifying pharmaceuticals in real-world settings. AI-based pill recognition models offer a promising solution, but the lack of comprehensive datasets hinders their development. Existing pill image datasets rarely capture real-world complexities such as overlapping pills, varied lighting, and occlusions. MEDISEG addresses this gap by providing instance segmentation annotations for 32 distinct pill types across 8262 images, encompassing diverse conditions from individual pill images to cluttered dosette boxes. We trained YOLOv8 and YOLOv9 on MEDISEG to demonstrate their usability, achieving mean average precision at IoU 0.5 of 99.5 percent on the 3-Pills subset and 80.1 percent on the 32-Pills subset. We further evaluate MEDISEG under a few-shot detection protocol, demonstrating that base training on MEDISEG significantly improves recognition of unseen pill classes in occluded multi-pill scenarios compared to existing datasets. These results highlight the dataset's ability not only to support robust supervised training but also to promote transferable representations under limited supervision, making it a valuable resource for developing and benchmarking AI-driven systems for medication safety.

Problem

• medication errors
• adverse drug events
• pill recognition
• instance segmentation
• real-world complexity

Innovation

• instance segmentation
• medication safety
• pill recognition
• few-shot detection
• MEDISEG dataset