MonoUNet: A Robust Tiny Neural Network for Automated Knee Cartilage Segmentation on Point-of-Care Ultrasound Devices
2026-04-09 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed MonoUNet, a very small and efficient deep learning model to automatically identify knee cartilage in ultrasound images taken by various portable devices. They designed MonoUNet with special features to better handle differences in image quality and appearance across devices. Tested on data from multiple machines and clinics, MonoUNet was more accurate and much faster than similar models, closely matching manual measurements of cartilage thickness and texture. This work shows that using particular image features helps compact models work well in challenging medical imaging tasks. The authors have also shared their code for others to use.
knee cartilageultrasound segmentationdeep learningU-Netpoint-of-care ultrasound (POCUS)local phase featuresmonogenic signalDice scoreintraclass correlation coefficientmodel robustness
Authors
Alvin Kimbowa, Arjun Parmar, Ibrahim Mujtaba, Will Wei, Maziar Badii, Matthew Harkey, David Liu, Ilker Hacihaliloglu
Abstract
Objective: To develop a robust and compact deep learning model for automated knee cartilage segmentation on point-of-care ultrasound (POCUS) devices. Methods: We propose MonoUNet, an ultra-compact U-Net consisting of (i) an aggressively reduced backbone with an asymmetric decoder, (ii) a trainable monogenic block that extracts multi-scale local phase features, and (iii) a gated feature injection mechanism that integrates these features into the encoder stages to reduce sensitivity to variations in ultrasound image appearance and improve robustness across devices. MonoUNet was evaluated on a multi-site, multi-device knee cartilage ultrasound dataset acquired using cart-based, portable, and handheld POCUS devices. Results: Overall, MonoUNet outperformed existing lightweight segmentation models, with average Dice scores ranging from 92.62% to 94.82% and mean average surface distance (MASD) values between 0.133 mm and 0.254 mm. MonoUNet reduces the number of parameters by 10x--700x and computational cost by 14x--2000x relative to existing lightweight models. MonoUNet cartilage outcomes showed excellent reliability and agreement with the manual outcomes: intraclass correlation coefficients (ICC$_{2,k})$=0.96 and bias=2.00% (0.047 mm) for average thickness, and ICC$_{2,k}$=0.99 and bias=0.80% (0.328 a.u.) for echo intensity. Conclusion: Incorporating trainable local phase features improves the robustness of highly compact neural networks for knee cartilage segmentation across varying acquisition settings and could support scalable ultrasound-based assessment and monitoring of knee osteoarthritis using POCUS devices. The code is publicly available at https://github.com/alvinkimbowa/monounet.