Related Experiment Video
Updated: May 27, 2025
06:24
Dynamic Contrast Enhanced Magnetic Resonance Imaging of an Orthotopic Pancreatic Cancer Mouse Model
Published on: April 18, 2015
15.1K
A deep learning detection method for pancreatic cystic neoplasm based on Mamba architecture
Junlong Dai1, Cong He1, Liang Jin2
1School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, China.
Journal of X-Ray Science and Technology
|February 20, 2025
Summary
The novel M-YOLO deep learning model significantly improves early detection of pancreatic cystic neoplasms (PCN) by combining Mamba and YOLOv10 architectures. This advancement aids in identifying complex tumor features for better patient outcomes.
Area of Science:
- Medical Imaging Analysis
- Artificial Intelligence in Healthcare
- Deep Learning for Oncology
Background:
- Early diagnosis of pancreatic cystic neoplasms (PCN) is critical for improving patient survival rates.
- Detecting PCNs is challenging due to their complex morphological features in medical images.
Purpose of the Study:
- To introduce M-YOLO, a novel deep learning model integrating Mamba architecture and YOLOv10.
- To enhance the accuracy and efficiency of PCN detection using advanced AI.
Main Methods:
- Developed M-YOLO (Mamba YOLOv10), a hybrid deep learning network.
- Leveraged Mamba's sequence modeling for contextual information in medical images.
- Utilized YOLOv10 for fast object detection to ensure clinical viability.
Main Results:
- M-YOLO achieved high performance metrics: 0.98 sensitivity, 0.92 specificity, 0.96 precision, 0.97 F1-score, 0.93 accuracy.
- The model demonstrated a mean average precision (mAP) of 0.96 at a 50% IoU threshold.
- Results were validated on a dataset from Changhai Hospital.
Conclusions:
- M-YOLO effectively enhances PCN identification performance.
- The model integrates Mamba's deep feature extraction with YOLOv10's fast localization.
- This integration offers a promising tool for early and accurate PCN diagnosis.
