SPOTnet: Weakly-Supervised Specular Reflection Detection and Removal for Minimal Invasive Surgery

Alam, Mahmood and Bilal, Muhammad and Khan, Ufaq and Cohen, Emil and Bapu, Deepa and Hajivayand, Amir M. and Khan, Kifayat Ullah and Akanbi, Lukman Adewale and Ajayi, Anuoluwapo and Kumar, Vikas and Xie, Yutong and Ali, Muhammad A. and Beggs, Andrew D. and Khan, Muhammad Haris and Ashraf, Shazad Q. (2026) SPOTnet: Weakly-Supervised Specular Reflection Detection and Removal for Minimal Invasive Surgery. npj Digital Surgery. ISSN 3091-2806 (In Press)

Text SPOTnet_Sept_2026.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Attribution. Download (20MB) | Request a copy

Abstract

Specular reflections are a common artefact in minimally invasive surgery, obscuring anatomical landmarks and degrading downstream computer vision systems relied upon for artificial intelligence-assisted surgical analysis. Existing approaches either require labour-intensive annotation or fail to generalise across heterogeneous surgical imaging. Here, we present SPOTnet (Specular Pseudo-label Optimised Training Network), a weakly-supervised framework for specular reflection detection and removal that requires no manual labelling. Our three-stage pipeline automatically builds a library of specular exemplars using hue–saturation–value (HSV) colour-space detection with morphological refinement, trains a lightweight U-Net on the resulting 500,000 patch-mask pairs, and integrates Large Mask Inpainting (LaMa) for restoration. Developed on TEMSET-24K (24,306 transanal endoscopic microsurgery frames), SPOTnet attains an IoU of 0.613, a recall of 0.740, and an E-measure of 0.963 on the annotated test set, outperforming recent methods including EndoSRR, Li et al., and UnReflectAnything, and demonstrates cross-dataset generalisation across four independent benchmarks without retraining. Specular removal improved downstream semantic segmentation (mean F1 +26.9%, p<0.01) on TEMSET-24K and polyp segmentation F1 from 0.938 to 0.956 on EndoScene. Prospective evaluation involving 13 surgeons (207 assessments) supported clinical acceptability, with 88.9% rating the processed video as trustworthy. SPOTnet is a scalable, annotation-free pre-processing step that improves downstream surgical computer vision.

Item Type:	Article
Dates:	Date Event 7 September 2026 Accepted
Uncontrolled Keywords:	Specular Reflection Detection, Endoscopic Video Inpainting, Weakly Supervised Learning, HSV Colour Space, U-Net
Subjects:	CAH02 - subjects allied to medicine > CAH02-05 - medical sciences > CAH02-05-03 - biomedical sciences (non-specific) CAH17 - business and management > CAH17-01 - business and management > CAH17-01-02 - business studies
Divisions:	Business School > Accountancy, Finance and Economics
Depositing User:	Gemma Tonks
Date Deposited:	23 Sep 2026 12:09
Last Modified:	23 Sep 2026 12:09
URI:	https://www.open-access.bcu.ac.uk/id/eprint/17251

Actions (login required)

View Item