Background and aims: Central nervous system is a crucial factor in chronic pain. There are only limited ways to evaluate animal pain in the brain. Manganese-Enhanced MRI is an attractive tool to evaluate animal brain objectively using MRI under anesthesia
Methods: After the animal ethic committee’s approval, twenty-five C57BL/6J mice were involved in this study. Eighteen mice were anaesthetized by pentobarbital with sevoflurane and received planter flap surgery as a chronic pain model. The scanning schedule was set 4 weeks after surgery and 2 days before the scanning day after which 15mg/kg manganese was injected intravenously for two days. MRI measurements were performed with an 11.7 T MR scanner, under isoflurane anesthesia, T1 weighted MR-images were acquired with a spin echo sequence. All the images were anatomically normalized and divided into 19 regions automatically. Furthermore, uptakes of manganese were analyzed by calculating the region-to-muscle uptake ratio. A support vector regression analysis was performed using MatLab. This is to calculate the pain score using signal intensity data in each animal brain. Average score was calculated.
Results: Support vector regression calculated the average score as 27.70 in 100mg/kg gabapentin administered group whereas 74.14 was the average score in 10mg/kg administered group respectively when the MRI signal intensity data labeled as 100 in group after surgery without analgesics and as 0 in naïve group.
Conclusions: This study revealed that we can quantify the effects of different dose of gabapentin to rodent chronic pain using Manganese-Enhanced MRI with support vector regression which is one kind of machine learning algorithms.

DATA DRIVEN QUANTIFICATION OF THE EFFECTS OF DIFFERENT DOSES OF GABAPENTIN FOR PAIN IN CHRONIC PHASE IN MICE USING MANGANESE-ENHANCED MRI WITH AI BASED ANALYSES