Researchers developed an AI algorithm to identify smuggled marine wildlife in airport luggage using CT scans with high accuracy.
Key Details
- 1Marine wildlife trafficking is valued in the billions annually and threatens ecosystems.
- 2Scientists used airport CT scanners to collect 298 scans of 3 key trafficked items: shark fins, seahorses, and sea cucumbers.
- 3A neural network was trained to detect these items within diverse luggage conditions, including common concealment tricks.
- 4The algorithm achieved 92% overall success: 95% for shark fins, 96% for seahorses, and 86% for sea cucumbers.
- 5False positive rate was 13% (lowest for shark fins at 2%, highest for seahorses at 9%).
- 6Study authors have disclosed employment by the company producing the X-ray system used.
Why It Matters
This research showcases how imaging AI can address non-medical, high-impact challenges by automating illicit wildlife detection in large-scale airport screening. It highlights potential for real-world adoption—and limitations—of AI in operational radiology infrastructure.
Source
EurekAlert
Related News
•EurekAlert
Broadband Optical Spectroscopy Enables Early NEC Detection in Preemies
Researchers successfully used a noninvasive broadband optical spectroscopy (BOS) device to detect necrotizing enterocolitis (NEC) early in premature infants.
•EurekAlert
AI Predicts Brain Tumor Risks from Routine Pathology Slides
Mayo Clinic developed an AI that analyzes routine pathology slides to classify meningiomas and predict recurrence risk, reducing dependency on advanced genetic testing.
•EurekAlert
UNM Researchers Use AI to Uncover Hidden Self-Harm Records in VA Data
Researchers develop a machine learning method to reveal self-harm history hidden in veterans' electronic medical records.