Abstract
Preoperative 3D segmentation of the tibiofemoral joint guides image-based Total Knee Arthroplasty (TKA) and supports the design of patient-specific implants. Automated 3D segmentation in TKA still remains challenging due to thin, low contrast cartilage layers between the tibiofemoral bone structures. In this paper, we propose Mamba Edge-Net
(MBEDNet), which integrates the local feature extraction of convolution with long dependency modelling of Mamba to effectively and efficiently segment tibiofemoral anatomies, with sharp contrast between bone and cartilage. MBEDNet is lightweight while achieving the state-of-the-art (SOTA) average Dice similarity coefficient (DSC), against architectures that require 1.3 to 5.5 times more parameters. We have also optimized the network to handle severe osteophytes of Kellgren-Lawrence (KL)
grade IV. Evaluating our model on the publicly available OAIZIB-CM knee magnetic resonance imaging (MRI) dataset, we achieved DSC scores of 98.46% and 98.64% on the femur and tibia respectively, and 89.03% and 86.29% on the femoral and tibial cartilage respectively. The code is available at https://github.com/Armaanmerc/MBEDNet
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/COLAS_017.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/forum?id=91LArGcPTD
BibTex
@InProceedings{PatArm_MBEDNet_MICCAISAT2026,
author = { Patnaik, Armaan AND Prasad, Ritanshu AND Hyagreev, Pranav AND B., Dharshan AND Maik, Vivek AND Lakshmanan, Manojkumar AND Sivaprakasam, Mohanasankar},
title = { { MBEDNet: A Mamba-CNN Network with Dual Edge-Guided Decoders for Segmentation of TibioFemoral Bone and Cartilage } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS pending},
month = {pending},
page = {pending}
}