Robust deep learning-based semantic organ segmentation in hyperspectral images

Seidlitz, Silvia; Sellner, Jan; Odenthal, Jan; Özdemir, Berkin; Studier-Fischer, Alexander; Knoedler, Samuel; Ayala, Leonardo; Adler, Tim; Kenngott, Hannes; Tizabi, Minu D.; Wagner, Martin; Nickel, Felix; Müller‐Stich, Beat P.; Maier-Hein, Lena

doi:10.1016/j.media.2022.102488

Cited by 35 publications

(17 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In RS-based architectures, the maximal kernel size is 3. As discovered in [ 37 ], larger sample sizes perform better. Probably it can be also applicable to kernel sizes.…”

Section: Discussionmentioning

confidence: 61%

Impact of Pre- and Post-Processing Steps for Supervised Classification of Colorectal Cancer in Hyperspectral Images

Tkachenko

Chalopin

Jansen-Winkeln

et al. 2023

Cancers

View full text Add to dashboard Cite

Background: Recent studies have shown that hyperspectral imaging (HSI) combined with neural networks can detect colorectal cancer. Usually, different pre-processing techniques (e.g., wavelength selection and scaling, smoothing, denoising) are analyzed in detail to achieve a well-trained network. The impact of post-processing was studied less. Methods: We tested the following methods: (1) Two pre-processing techniques (Standardization and Normalization), with (2) Two 3D-CNN models: Inception-based and RemoteSensing (RS)-based, with (3) Two post-processing algorithms based on median filter: one applies a median filter to a raw predictions map, the other applies the filter to the predictions map after adopting a discrimination threshold. These approaches were evaluated on a dataset that contains ex vivo hyperspectral (HS) colorectal cancer records of 56 patients. Results: (1) Inception-based models perform better than RS-based, with the best results being 92% sensitivity and 94% specificity; (2) Inception-based models perform better with Normalization, RS-based with Standardization; (3) Our outcomes show that the post-processing step improves sensitivity and specificity by 6.6% in total. It was also found that both post-processing algorithms have the same effect, and this behavior was explained. Conclusion: HSI combined with tissue classification algorithms is a promising diagnostic approach whose performance can be additionally improved by the application of the right combination of pre- and post-processing.

show abstract

“…In RS-based architectures, the maximal kernel size is 3. As discovered in [ 37 ], larger sample sizes perform better. Probably it can be also applicable to kernel sizes.…”

Section: Discussionmentioning

confidence: 61%

Impact of Pre- and Post-Processing Steps for Supervised Classification of Colorectal Cancer in Hyperspectral Images

Tkachenko

Chalopin

Jansen-Winkeln

et al. 2023

Cancers

View full text Add to dashboard Cite

show abstract

“…Hyperspectral Imaging (HSI) is a fast and noninvasive and nonionizing monitoring technique based on spectrometric tissue analysis (88)(89)(90). The target tissue is illuminated by halogen lamps and remitted light is detected in a wavelength spectrum from visual to near-infrared light (380-1000 nm) (91).…”

Section: Hyperspectral Imagingmentioning

confidence: 99%

Postoperative free flap monitoring in reconstructive surgery—man or machine?

Knoedler

Hoch²,

Huelsboemer³

et al. 2023

Front. Surg.

View full text Add to dashboard Cite

Free tissue transfer is widely used for the reconstruction of complex tissue defects. The survival of free flaps depends on the patency and integrity of the microvascular anastomosis. Accordingly, the early detection of vascular comprise and prompt intervention are indispensable to increase flap survival rates. Such monitoring strategies are commonly integrated into the perioperative algorithm, with clinical examination still being considered the gold standard for routine free flap monitoring. Despite its widespread acceptance as state of the art, the clinical examination also has its pitfalls, such as the limited applicability in buried flaps and the risk of poor interrater agreement due to inconsistent flap (failure) appearances. To compensate for these shortcomings, a plethora of alternative monitoring tools have been proposed in recent years, each of them with inherent strengths and limitations. Given the ongoing demographic change, the number of older patients requiring free flap reconstruction, e.g., after cancer resection, is rising. Yet, age-related morphologic changes may complicate the free flap evaluation in elderly patients and delay the prompt detection of clinical signs of flap compromise. In this review, we provide an overview of currently available and employed methods for free flap monitoring, with a special focus on elderly patients and how senescence may impact standard free flap monitoring strategies.

show abstract

“…Similarly, the segmentation of organs [128], [227], [228] and tissues is also well studied task in RAS [171]. The Mask-RCN and CNN based YOLO, U-Net, TernausNet, LinkNet, and SegNet are applied on famous EndoVis Challenge and in-house datasets.…”

Section: E Othersmentioning

confidence: 99%

Deep Learning Based Image Processing for Robot Assisted Surgery: A Systematic Literature Survey

et al. 2022

View full text Add to dashboard Cite

The recent advancements in the surging field of Deep Learning (DL) have revolutionized every sphere of life, and the healthcare domain is no exception. The enormous success of DL models, particularly with image data, has led to the development of image-guided Robot Assisted Surgery (RAS) systems. By and large, the number of studies concerning image-driven computer assisted surgical systems using DL has increased exponentially. Additionally, the contemporary availability of surgical datasets has also boosted the DL applications in RAS. Inspired by the latest trends and contributions in surgery, this literature survey presents a summarized analysis of recent innovations of DL in image-guided RAS systems. After a thorough review, a sum of 184 articles are selected and grouped into four categories, based on the literature and the relevancy of the task in the articles, comprising 1) Surgical Tools, 2) Surgical Processes, 3) Surgical Surveillance, and 4) Surgical Performance. The survey also discusses publicly available surgical datasets and highlights the basics of the DL models. Furthermore, the legal, ethical, and technological challenges together with the intuitive predictions and recommendations related to the autonomous RAS systems are also presented. The study reveals that Convolutional Neural Network (CNN) is most widely adopted architecture, whereas, JIGSAWS is most employed dataset in RAS. The study suggests fusing kinematic data along with image data, which produces better accuracy and precision, particularly in gesture and trajectory segmentation tasks. Additionally, CNN and Long Short Term Memory (LSTM) networks have shown remarkable performance, however, authors recommend employing these gigantic architectures only when simpler models have failed to produce satisfactory results. The simpler models, despite their limitations, are time and cost effective and yield considerable outcomes even on the smaller datasets.

show abstract

Robust deep learning-based semantic organ segmentation in hyperspectral images

Cited by 35 publications

References 45 publications

Impact of Pre- and Post-Processing Steps for Supervised Classification of Colorectal Cancer in Hyperspectral Images

Impact of Pre- and Post-Processing Steps for Supervised Classification of Colorectal Cancer in Hyperspectral Images

Postoperative free flap monitoring in reconstructive surgery—man or machine?

Deep Learning Based Image Processing for Robot Assisted Surgery: A Systematic Literature Survey

Contact Info

Product

Resources

About