Dataset Information

Adaptive Boosting (AdaBoost)-based multiwavelength spatial frequency domain imaging and characterization for ex vivo human colorectal tissue assessment.

ABSTRACT: The current gold standard diagnostic test for colorectal cancer remains histological inspections of endoluminal neoplasia in biopsy specimens. However, biopsy site selection requires visual inspection of the bowel, typically with a white-light endoscope. Therefore, this technique is poorly suited to detect small or innocuous-appearing lesions. We hypothesize that an alternative modality-multiwavelength spatial frequency domain imaging (SFDI)-would be able to differentiate various colorectal neoplasia from normal tissue. In this ex vivo study of human colorectal tissues, we report the optical absorption and scattering signatures of normal, adenomatous polyp and cancer specimens. An abnormal vs. normal adaptive boosting (AdaBoost) classifier is trained to dichotomize tissue based on SFDI imaging characteristics, and an area under the receiver operating characteristic (ROC) curve (AUC) of 0.95 is achieved. We conclude that AdaBoost-based multiwavelength SFDI can differentiate abnormal from normal colorectal tissues, potentially improving endoluminal screening of the distal gastrointestinal tract in the future.

SUBMITTER: Li S

PROVIDER: S-EPMC7593835 | biostudies-literature | 2020 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Adaptive Boosting (AdaBoost)-based multiwavelength spatial frequency domain imaging and characterization for ex vivo human colorectal tissue assessment.

Li Shuying S Zeng Yifeng Y Chapman William C WC Erfanzadeh Mohsen M Nandy Sreyankar S Mutch Matthew M Zhu Quing Q

Journal of biophotonics 20200325 6

The current gold standard diagnostic test for colorectal cancer remains histological inspections of endoluminal neoplasia in biopsy specimens. However, biopsy site selection requires visual inspection of the bowel, typically with a white-light endoscope. Therefore, this technique is poorly suited to detect small or innocuous-appearing lesions. We hypothesize that an alternative modality-multiwavelength spatial frequency domain imaging (SFDI)-would be able to differentiate various colorectal neop ...[more]

PMID: 32125775

Similar Datasets

Project description:Spatial frequency domain (SFD) measurement allows rapid and non-contact wide-field imaging of the tissue optical properties, thus has become a potential tool for assessing physiological parameters and therapeutic responses during photodynamic therapy of skin diseases. The conventional SFD measurement requires a reference measurement within the same experimental scenario as that for a test one to calibrate mismatch between the real measurements and the model predictions. Due to the individual physical and geometrical differences among different tissues, organs and patients, an ideal reference measurement might be unavailable in clinical trials. To address this problem, we present a reference-free SFD determination of absorption coefficient that is based on the modulation transfer function (MTF) characterization.Instead of the absolute amplitude that is used in the conventional SFD approaches, we herein employ the MTF to characterize the propagation of the modulated lights in tissues. With such a dimensionless relative quantity, the measurements can be naturally corresponded to the model predictions without calibrating the illumination intensity. By constructing a three-dimensional database that portrays the MTF as a function of the optical properties (both the absorption coefficient μ a and the reduced scattering coefficient [Formula: see text]) and the spatial frequency, a look-up table approach or a least-square curve-fitting method is readily applied to recover the absorption coefficient from a single frequency or multiple frequencies, respectively.Simulation studies have verified the feasibility of the proposed reference-free method and evaluated its accuracy in the absorption recovery. Experimental validations have been performed on homogeneous tissue-mimicking phantoms with μ a ranging from 0.01 to 0.07 mm-1 and [Formula: see text] = 1.0 or 2.0 mm-1. The results have shown maximum errors of 4.86 and 7% for [Formula: see text] = 1.0 mm-1 and [Formula: see text] = 2.0 mm-1, respectively. We have also presented quantitative ex vivo imaging of human lung cancer in a subcutaneous xenograft mouse model for further validation, and observed high absorption contrast in the tumor region.The proposed method can be applied to the rapid and accurate determination of the absorption coefficient, and better yet, in a reference-free way. We believe this reference-free strategy will facilitate the clinical translation of the SFD measurement to achieve enhanced intraoperative hemodynamic monitoring and personalized treatment planning in photodynamic therapy.

Dataset Information

Adaptive Boosting (AdaBoost)-based multiwavelength spatial frequency domain imaging and characterization for ex vivo human colorectal tissue assessment.

Publications

Adaptive Boosting (AdaBoost)-based multiwavelength spatial frequency domain imaging and characterization for ex vivo human colorectal tissue assessment.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets