Intelligent Identification of Early Esophageal Cancer by Band-Selective Hyperspectral Imaging

Abstract

In this study, the combination of hyperspectral imaging (HSI) technology and band selection was coupled with color reproduction. The white-light images (WLIs) were simulated as narrow-band endoscopic images (NBIs). As a result, the blood vessel features in the endoscopic image became more noticeable, and the prediction performance was improved. In addition, a single-shot multi-box detector model for predicting the stage and location of esophageal cancer was developed to evaluate the results. A total of 1780 esophageal cancer images, including 845 WLIs and 935 NBIs, were used in this study. The images were divided into three stages based on the pathological features of esophageal cancer: normal, dysplasia, and squamous cell carcinoma. The results showed that the mean average precision (mAP) reached 80% in WLIs, 85% in NBIs, and 84% in HSI images. This study′s results showed that HSI has more spectral features than white-light imagery, and it improves accuracy by about 5% and matches the results of NBI predictions.

Analysis of optical absorption of photoaged human skin using a high‐frequency illumination microscopy analysis system

“…Recently, many efforts have been put into developing artificial intelligence (AI) for image recognition to make diagnoses using appearance data. [32][33][34] They demonstrate the potential of hyperspectral imaging and highlight the importance of using machine-learning and deep-learning algorithms to analyse medical images. Their aim is to develop algorithms and models to process and analyse medical images.…”

“…No one would deny that appearance is a key indicator in diagnosis. Recently, many efforts have been put into developing artificial intelligence (AI) for image recognition to make diagnoses using appearance data 32–34 . They demonstrate the potential of hyperspectral imaging and highlight the importance of using machine‐learning and deep‐learning algorithms to analyse medical images.…”

Hyperspectral imaging for early diagnosis of diseases: A review

“…T.‐J. Tsai et al, 2022 apply hyperspectral band selection for an early identification of esophagal cancer. The results obtained showcase an increase in accuracy by 5% due to enhancement of characteristics in white light imaging (WLIs).…”

Automatic Counterfeit Currency Detection Using a Novel Snapshot Hyperspectral Imaging Algorithm

“…HSI acquires the spectrum for each pixel in an image [ 18 , 19 , 20 , 21 ]. It has been used in many applications, such as cancer detection [ 22 , 23 , 24 , 25 ], air pollution monitoring [ 26 , 27 ], nanostructure identification [ 28 , 29 , 30 , 31 ], aerospace [ 32 , 33 , 34 ], food quality maintenance [ 35 ], verification [ 36 , 37 , 38 ], military [ 39 ], remote sensing [ 40 , 41 , 42 ], and agriculture [ 43 ].…”