Hard Frame Detection and Online Mapping for Surgical Phase Recognition

Surgical phase recognition is an important topic of Computer Assisted Surgery (CAS) systems. In the complicated surgical procedures, there are lots of hard frames that have indistinguishable visual features but are assigned with different labels. Prior works try to classify hard frames along with other simple frames indiscriminately, which causes various problems. Different from previous approaches, we take hard frames as mislabeled samples and find them in the training set via data cleansing strategy. Then, we propose an Online Hard Frame Mapper (OHFM) to handle the detected hard frames separately. We evaluate our solution on the M2CAI16 Workflow Challenge dataset and the Cholec80 dataset and achieve superior results. (The code is available at https://github.com/ChinaYi/miccai19).

[1]  Gregory D. Hager,et al.  Surgical Gesture Segmentation and Recognition , 2013, MICCAI.

[2]  Chi-Wing Fu,et al.  SV-RCNet: Workflow Recognition From Surgical Videos Using Recurrent Convolutional Network , 2018, IEEE Transactions on Medical Imaging.

[3]  Nathalie Bricon-Souf,et al.  Context awareness in health care: A review , 2007, Int. J. Medical Informatics.

[4]  Nassir Navab,et al.  Modeling and Segmentation of Surgical Workflow from Laparoscopic Video , 2010, MICCAI.

[5]  Pierre Jannin,et al.  A Framework for the Recognition of High-Level Surgical Tasks From Video Images for Cataract Surgeries , 2012, IEEE Transactions on Biomedical Engineering.

[6]  M. Verleysen,et al.  Classification in the Presence of Label Noise: A Survey , 2014, IEEE Transactions on Neural Networks and Learning Systems.

[7]  Nassir Navab,et al.  On-line Recognition of Surgical Activity for Monitoring in the Operating Room , 2008, AAAI.

[8]  Nassir Navab,et al.  Statistical modeling and recognition of surgical workflow , 2012, Medical Image Anal..

[9]  Gregory D. Hager,et al.  Automatic Detection and Segmentation of Robot-Assisted Surgical Motions , 2005, MICCAI.

[10]  Peter Fu-Ming Hu,et al.  Real-Time Identification of Operating Room State from Video , 2007, AAAI.

[11]  Shifu Chen,et al.  Identifying and Correcting Mislabeled Training Instances , 2007, Future Generation Communication and Networking (FGCN 2007).

[12]  Matthieu Cord,et al.  M2CAI Workflow Challenge: Convolutional Neural Networks with Time Smoothing and Hidden Markov Model for Video Frames Classification , 2016, ArXiv.

[13]  Pierre Jannin,et al.  Surgical Phases Detection from Microscope Videos by Combining SVM and HMM , 2010, MCV.

[14]  André Carlos Ponce de Leon Ferreira de Carvalho,et al.  Use of Classification Algorithms in Noise Detection and Elimination , 2009, HAIS.

[15]  Saso Dzeroski,et al.  Noise detection and elimination in data preprocessing: Experiments in medical domains , 2000, Appl. Artif. Intell..

[16]  Andru Putra Twinanda,et al.  Single- and Multi-Task Architectures for Surgical Workflow Challenge at M2CAI 2016 , 2016, ArXiv.

[17]  Andru Putra Twinanda,et al.  EndoNet: A Deep Architecture for Recognition Tasks on Laparoscopic Videos , 2016, IEEE Transactions on Medical Imaging.