A French retrospective study evaluated two AI tools: Pixyl.Neuro.MS (Pixyl) and Jazz (AI Medical) for detecting new T2/FLAIR lesions in 35 multiple sclerosis patients with high lesion loads. The aim was to assess whether AI-assisted readings improve lesion detection compared to standard radiology reports.
Two radiologists (15 and 4 years’ experience) independently reviewed longitudinal 3D FLAIR MRIs. Ground truth confirmed 21 new lesions in 9 patients, whereas standard reports identified 8 lesions in 2 patients (22.2% sensitivity).
Pixyl.Neuro.MS automatically segmented and labeled lesions, generating color-coded overlays and structured reports. It achieved 100% sensitivity for both readers but produced 61 false positives. Jazz did not detect or label lesions; instead, it co-registered scans and enabled synchronized image comparison. Sensitivity with Jazz reflected radiologist performance using the software, ranging from 77.8% to 88.9%, with 100% specificity. Average reading times were 2:46 with Pixyl and 3:33 with Jazz. Inter-reader agreement was high for both tools (Pixyl r = 0.73; Jazz r = 0.75; p < 0.001).
The study shows that both AI tools enhance new lesion detection in MS follow-up imaging. While Pixyl may favor automation and sensitivity, and Jazz supports radiologist-guided precision, both require expert validation to ensure clinical reliability.
Evaluation of two AI techniques for the detection of new T2/FLAIR lesions in the follow-up of multiple sclerosis patients
Frontiers in Neurology, 2025
Abstract
Background
Multiple sclerosis is an inflammatory demyelinating disease of the CNS. Annual MRI exams are crucial for disease monitoring. Interpreting high T2/FLAIR lesion loads can be laborious. AI aids in lesion detection, and choosing between different solutions can be challenging.
Aim
This study compares two distinct software, Pixyl.Neuro.MS® and Jazz®, to assess their performance in T2/FLAIR lesion detection between two-time points.
Methods
Retrospective analysis included follow-up MRIs from 35 MS patients. Pixyl.Neuro.MS® automatically segments and classifies lesions. Jazz® automates the reading process and image display. Two readers (15 and 4 years of experience) conducted radiological analysis, followed by AI-assisted readings. A number of new lesions (NL) and reading times were recorded, with ground truth (GT) established by consensus. AI-detected lesions were classified as true (TP) and false positives (FP). Statistical analysis used SPSS (p < 0.05).
Results
Pixyl.Neuro.MS® readings averaged 2 min 46 s ± 1 min 4 s while using Jazz® 3 min 33 s ± 2 min 24 s. Over 50% of the population had a high lesion load (>20 lesions). Both software significantly improved NL detection (p < 0.01 for both), revealing them in more patients than standard readings. Standard reports found 8 NL in 2 patients, while AI-assisted readings detected at least 17 TP in 7 patients and rejected 61 FP lesions. GT detected 21 lesions in 19 patients.
Conclusion
Both AI software have been found to enhance NL detection in MS patients, outperforming standard methods. These tools offer crucial advantages for accurate disease monitoring.
