Optimizing algorithm development for tissue classification in colorectal cancer based on diffuse reflectance spectra

Abstract

Diffuse reflectance spectroscopy can be used in colorectal cancer surgery for tissue classification. The main challenge in the classification task is to separate healthy colorectal wall from tumor tissue. In this study, four normalization techniques, four feature extraction methods and five classifiers are applied to nine datasets, to obtain the optimal method to separate spectra measured on healthy colorectal wall from spectra measured on tumor tissue. All results are compared to the use of the entire non-normalized spectra. It is found that the most optimal classification approach is to apply a feature extraction method on non-normalized spectra combined with support vector machine or neural network classifier.

Evaluation of wavelength ranges and tissue depth probed by diffuse reflectance spectroscopy for colorectal cancer detection

“…Relevant biochemical and structural differences for CRC detection. From a clinical perspective, biomolecular changes probed by optical techniques can potentially obviate the need for multiple biopsies or polypectomies of normal mucosa as well as identify sessile serrated polyps, which may be difficult to recognize during colonoscopy at times [7][8][9][10][11][12][13][14][15][16][17][18][19][20] . The potential of optical spectroscopy for colorectal cancer (CRC) detection in ex vivo specimens or in vivo during colonoscopy has been evaluated for superficial tissues (small SDD probes) in several wavelength ranges [39][40][41][42][43][44][45][46][47][48][49][50] .…”

Layer thickness prediction and tissue classification in two-layered tissue structures using diffuse reflectance spectroscopy

“…In colorectal cancer surgery, the tumor is assessed from outside the lumen and should be distinguished from fat and healthy colorectal wall, in contrast to colonoscopy where the tumor is assessed from the inside of the colorectal lumen. Baltussen et al examined several options to optimize the classification of healthy colorectal wall versus tumor tissue and showed that tissue optical features, in which tissue constituents such as water and fat content were determined per spectrum using a fit algorithm, improved classification results the most 53 . For more complex situations, such as discriminating tumor versus muscle tissue, other discriminative features using more advanced feature extraction techniques can be used.…”

Real-time tracking of a diffuse reflectance spectroscopy probe used to aid histological validation of margin assessment in upper gastrointestinal cancer resection surgery

“…Binary classification into normal and tumor tissue was performed using various supervised machine learning classifiers, such as linear support vector machine (SVM), multi-layer perceptron (MLP), light gradient boosting machine (LGBM), and extreme gradient boosting (XGB). 26 …”