Identification of Colorectal Cancer Using Near-Infrared Spectroscopy and Adaboost with Decision Stump

Chen, Hui; Lin, Zan; Mo, Lin; Tan, Chao

doi:10.1080/00032719.2017.1310880

Cited by 12 publications

(15 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Usefulness of the extended wavelength range. Based on the classification performance on the wavelength ranges of previous studies [39][40][41][42][43][44][45][46][47][48][49][50] , we analyzed the potential benefit of using the extended wavelength for colorectal cancer (CRC) detection. In order to be consistent with the findings of previous studies, this analysis consisted of the comparison of the sensitivity, specificity, accuracy and AUC using the small SDD probe (Table 4).…”

Section: Resultsmentioning

confidence: 99%

“…Comparison with previous studies. The classification performance for colorectal cancer detection achieved in this study was compared to similar studies investigating DRS, elastic scattering, near-infrared spectroscopy (reflectance modality) and hyperspectral imaging [39][40][41][42][43][44][45][46][47][48][49][50] . This comparison involved listing the type of tissue evaluation (in vivo or ex vivo), number of patients, wavelength range, source-to-detector distance, types of tissue analyzed, number of analyzed spectra, tissue types used for classification, and diagnostic performance metrics (sensitivity, specificity, accuracy, and AUC).…”

mentioning

confidence: 99%

See 1 more Smart Citation

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

Nogueira

Maryam

Amissah

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Colorectal cancer (CRC) is the third most common type of cancer worldwide and the second most deadly. Recent research efforts have focused on developing non-invasive techniques for CRC detection. In this study, we evaluated the diagnostic capabilities of diffuse reflectance spectroscopy (DRS) for CRC detection by building 6 classification models based on support vector machines (SVMs). Our dataset consists of 2889 diffuse reflectance spectra collected from freshly excised ex vivo tissues of 47 patients over wavelengths ranging from 350 and 1919 nm with source-detector distances of 630-µm and 2500-µm to probe different depths. Quadratic SVMs were used and performance was evaluated using twofold cross-validation on 10 iterations of randomized training and test sets. We achieved (93.5 ± 2.4)% sensitivity, (94.0 ± 1.7)% specificity AUC by probing the superficial colorectal tissue and (96.1 ± 1.8)% sensitivity, (95.7 ± 0.6)% specificity AUC by sampling deeper tissue layers. To the best of our knowledge, this is the first DRS study to investigate the potential of probing deeper tissue layers using larger SDD probes for CRC detection in the luminal wall. The data analysis showed that using a broader spectrum and longer near-infrared wavelengths can improve the diagnostic accuracy of CRC as well as probing deeper tissue layers.

show abstract

Section: Resultsmentioning

confidence: 99%

mentioning

confidence: 99%

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

Nogueira

Maryam

Amissah

et al. 2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…8,17,18 Raman, near-infrared, and fluorescence-based spectroscopic modalities have been established with varying specificity and sensitivity in distinguishing benign and malignant colonic tissue. 19,20 In work presented herein, various spectroscopic modalities have been evaluated in their ability to detect CRC tissue in an ex vivo setting. The results of this pilot study show a potential advantage of spectroscopic modalities and should be analyzed in further clinical studies.…”

Section: Discussionmentioning

confidence: 99%

Synergy of Fluorescence and Near-Infrared Spectroscopy in Detection of Colorectal Cancer

Ehlen

Zabaryło

Speichinger

et al. 2019

Journal of Surgical Research

View full text Add to dashboard Cite

Magnetic resonance spectroscopy a b s t r a c tBackground: Colorectal cancer is one of the most common malignancies worldwide. There is an urgent need for simple and fast methods to improve tumor detection in the diagnostic and intraoperative setting to avoid complications and provide objective information in distinguishing malignant and benign colorectal tissue. Optical spectroscopy methods have recently shown a great potential for this discrimination in different organs. Materials and methods:In this pilot study, fluorescence emission spectra (excitation: 473 nm) and diffuse reflectance spectra (DRS) of normal and tumor tissues from resected colorectal cancer specimen were measured using fiber optical probes in an ex vivo setting, and the data were subjected to multivariate analysis.Results: Substantial spectral differences were found in the fluorescence and DRS spectra of colorectal cancer tissue in comparison to benign tissue. The diagnostic potential of a multimode optical system combining both spectroscopic methods was investigated by mathematical combination. Compared with the individual techniques, a higher sensitivity of the joint DRS-fluorescence optical system in the discrimination between malignant and benign colorectal tissue could be observed. Conclusions:In the pilot study presented herein, a quick and reliable method to differentiate malignant and benign colorectal tissue ex vivo with different spectroscopic techniques using spectral fiber probes could be established. Joint fluorescence and near-infrared spectroscopy had a higher sensitivity in tissue discrimination and showed to be a promising combination of two spectroscopic methods. Further studies using the synergic effect of fluorescence and DRS spectroscopy are needed to transfer these findings into the in vivo situation. ª

show abstract

“…A decision stump (DS) is an algorithm of a onetier decision tree with only an internal node (root) linked directly with terminal nodes (leaves) [44]. The DS forecasts can be based on a single input value and can be refereed a one-rule.…”

Section: Decision Stumpmentioning

confidence: 99%

“…It is a simplification derived from Adaptive Boosting. The AdaBoost ensemble is constructed by combining weak learners [44] , the weak learner (e.g., lets us use k-NN), and number of cycles T. 2) Initialization: The weights of training samples are initialized:…”

Section: Boosting (Adaboost) Ensemble Algorithmmentioning

confidence: 99%

The Detection of Counterfeit Banknotes Using Ensemble Learning Techniques of AdaBoost and Voting

Khairy¹,

Hussein²,

Alrikabi³

2021

IJIES

View full text Add to dashboard Cite

The movement of cash flow transactions by either electronic channels or physically created openings for the influx of counterfeit banknotes in financial markets. Aided by global economic integration and expanding international trade, attention must be geared at robust techniques for the recognition and detection of counterfeit banknotes. This paper presents ensemble learning algorithms for banknotes detection. The AdaBoost and voting ensemble are deployed in combination with machine learning algorithms. Improved detection accuracies are produced by the ensemble methods. Simulation results certify that the ensemble models of AdaBoost and voting provided accuracies of up to 100% for counterfeit banknotes.

show abstract

Identification of Colorectal Cancer Using Near-Infrared Spectroscopy and Adaboost with Decision Stump

Cited by 12 publications

References 21 publications

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

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

Synergy of Fluorescence and Near-Infrared Spectroscopy in Detection of Colorectal Cancer

The Detection of Counterfeit Banknotes Using Ensemble Learning Techniques of AdaBoost and Voting

Contact Info

Product

Resources

About