-
Semantic segmentation of surgical hyperspectral images under geometric domain shifts
Authors:
Jan Sellner,
Silvia Seidlitz,
Alexander Studier-Fischer,
Alessandro Motta,
Berkin Özdemir,
Beat Peter Müller-Stich,
Felix Nickel,
Lena Maier-Hein
Abstract:
Robust semantic segmentation of intraoperative image data could pave the way for automatic surgical scene understanding and autonomous robotic surgery. Geometric domain shifts, however, although common in real-world open surgeries due to variations in surgical procedures or situs occlusions, remain a topic largely unaddressed in the field. To address this gap in the literature, we (1) present the…
▽ More
Robust semantic segmentation of intraoperative image data could pave the way for automatic surgical scene understanding and autonomous robotic surgery. Geometric domain shifts, however, although common in real-world open surgeries due to variations in surgical procedures or situs occlusions, remain a topic largely unaddressed in the field. To address this gap in the literature, we (1) present the first analysis of state-of-the-art (SOA) semantic segmentation networks in the presence of geometric out-of-distribution (OOD) data, and (2) address generalizability with a dedicated augmentation technique termed "Organ Transplantation" that we adapted from the general computer vision community. According to a comprehensive validation on six different OOD data sets comprising 600 RGB and hyperspectral imaging (HSI) cubes from 33 pigs semantically annotated with 19 classes, we demonstrate a large performance drop of SOA organ segmentation networks applied to geometric OOD data. Surprisingly, this holds true not only for conventional RGB data (drop of Dice similarity coefficient (DSC) by 46 %) but also for HSI data (drop by 45 %), despite the latter's rich information content per pixel. Using our augmentation scheme improves on the SOA DSC by up to 67 % (RGB) and 90 % (HSI) and renders performance on par with in-distribution performance on real OOD test data. The simplicity and effectiveness of our augmentation scheme makes it a valuable network-independent tool for addressing geometric domain shifts in semantic scene segmentation of intraoperative data. Our code and pre-trained models are available at https://github.com/IMSY-DKFZ/htc.
△ Less
Submitted 17 September, 2023; v1 submitted 20 March, 2023;
originally announced March 2023.
-
Unsupervised Domain Transfer with Conditional Invertible Neural Networks
Authors:
Kris K. Dreher,
Leonardo Ayala,
Melanie Schellenberg,
Marco Hübner,
Jan-Hinrich Nölke,
Tim J. Adler,
Silvia Seidlitz,
Jan Sellner,
Alexander Studier-Fischer,
Janek Gröhl,
Felix Nickel,
Ullrich Köthe,
Alexander Seitel,
Lena Maier-Hein
Abstract:
Synthetic medical image generation has evolved as a key technique for neural network training and validation. A core challenge, however, remains in the domain gap between simulations and real data. While deep learning-based domain transfer using Cycle Generative Adversarial Networks and similar architectures has led to substantial progress in the field, there are use cases in which state-of-the-ar…
▽ More
Synthetic medical image generation has evolved as a key technique for neural network training and validation. A core challenge, however, remains in the domain gap between simulations and real data. While deep learning-based domain transfer using Cycle Generative Adversarial Networks and similar architectures has led to substantial progress in the field, there are use cases in which state-of-the-art approaches still fail to generate training images that produce convincing results on relevant downstream tasks. Here, we address this issue with a domain transfer approach based on conditional invertible neural networks (cINNs). As a particular advantage, our method inherently guarantees cycle consistency through its invertible architecture, and network training can efficiently be conducted with maximum likelihood training. To showcase our method's generic applicability, we apply it to two spectral imaging modalities at different scales, namely hyperspectral imaging (pixel-level) and photoacoustic tomography (image-level). According to comprehensive experiments, our method enables the generation of realistic spectral data and outperforms the state of the art on two downstream classification tasks (binary and multi-class). cINN-based domain transfer could thus evolve as an important method for realistic synthetic data generation in the field of spectral imaging and beyond.
△ Less
Submitted 17 March, 2023;
originally announced March 2023.
-
Robust deep learning-based semantic organ segmentation in hyperspectral images
Authors:
Silvia Seidlitz,
Jan Sellner,
Jan Odenthal,
Berkin Özdemir,
Alexander Studier-Fischer,
Samuel Knödler,
Leonardo Ayala,
Tim J. Adler,
Hannes G. Kenngott,
Minu Tizabi,
Martin Wagner,
Felix Nickel,
Beat P. Müller-Stich,
Lena Maier-Hein
Abstract:
Semantic image segmentation is an important prerequisite for context-awareness and autonomous robotics in surgery. The state of the art has focused on conventional RGB video data acquired during minimally invasive surgery, but full-scene semantic segmentation based on spectral imaging data and obtained during open surgery has received almost no attention to date. To address this gap in the literat…
▽ More
Semantic image segmentation is an important prerequisite for context-awareness and autonomous robotics in surgery. The state of the art has focused on conventional RGB video data acquired during minimally invasive surgery, but full-scene semantic segmentation based on spectral imaging data and obtained during open surgery has received almost no attention to date. To address this gap in the literature, we are investigating the following research questions based on hyperspectral imaging (HSI) data of pigs acquired in an open surgery setting: (1) What is an adequate representation of HSI data for neural network-based fully automated organ segmentation, especially with respect to the spatial granularity of the data (pixels vs. superpixels vs. patches vs. full images)? (2) Is there a benefit of using HSI data compared to other modalities, namely RGB data and processed HSI data (e.g. tissue parameters like oxygenation), when performing semantic organ segmentation? According to a comprehensive validation study based on 506 HSI images from 20 pigs, annotated with a total of 19 classes, deep learning-based segmentation performance increases, consistently across modalities, with the spatial context of the input data. Unprocessed HSI data offers an advantage over RGB data or processed data from the camera provider, with the advantage increasing with decreasing size of the input to the neural network. Maximum performance (HSI applied to whole images) yielded a mean DSC of 0.90 ((standard deviation (SD)) 0.04), which is in the range of the inter-rater variability (DSC of 0.89 ((standard deviation (SD)) 0.07)). We conclude that HSI could become a powerful image modality for fully-automatic surgical scene understanding with many advantages over traditional imaging, including the ability to recover additional functional tissue information. Code and pre-trained models: https://github.com/IMSY-DKFZ/htc.
△ Less
Submitted 10 July, 2022; v1 submitted 9 November, 2021;
originally announced November 2021.
-
Machine learning-based analysis of hyperspectral images for automated sepsis diagnosis
Authors:
Maximilian Dietrich,
Silvia Seidlitz,
Nicholas Schreck,
Manuel Wiesenfarth,
Patrick Godau,
Minu Tizabi,
Jan Sellner,
Sebastian Marx,
Samuel Knödler,
Michael M. Allers,
Leonardo Ayala,
Karsten Schmidt,
Thorsten Brenner,
Alexander Studier-Fischer,
Felix Nickel,
Beat P. Müller-Stich,
Annette Kopp-Schneider,
Markus A. Weigand,
Lena Maier-Hein
Abstract:
Sepsis is a leading cause of mortality and critical illness worldwide. While robust biomarkers for early diagnosis are still missing, recent work indicates that hyperspectral imaging (HSI) has the potential to overcome this bottleneck by monitoring microcirculatory alterations. Automated machine learning-based diagnosis of sepsis based on HSI data, however, has not been explored to date. Given thi…
▽ More
Sepsis is a leading cause of mortality and critical illness worldwide. While robust biomarkers for early diagnosis are still missing, recent work indicates that hyperspectral imaging (HSI) has the potential to overcome this bottleneck by monitoring microcirculatory alterations. Automated machine learning-based diagnosis of sepsis based on HSI data, however, has not been explored to date. Given this gap in the literature, we leveraged an existing data set to (1) investigate whether HSI-based automated diagnosis of sepsis is possible and (2) put forth a list of possible confounders relevant for HSI-based tissue classification. While we were able to classify sepsis with an accuracy of over $98\,\%$ using the existing data, our research also revealed several subject-, therapy- and imaging-related confounders that may lead to an overestimation of algorithm performance when not balanced across the patient groups. We conclude that further prospective studies, carefully designed with respect to these confounders, are necessary to confirm the preliminary results obtained in this study.
△ Less
Submitted 15 June, 2021;
originally announced June 2021.