This study uses hyperspectral imaging (HSI) and a deep learning diagnosis model that can identify the stage of esophageal cancer and mark the locations. This model simulates the spectrum data from the image using an algorithm developed in this study which is combined with deep learning for the classification and diagnosis of esophageal cancer using a single-shot multibox detector (SSD)-based identification system. Some 155 white-light endoscopic images and 153 narrow-band endoscopic images of esophageal cancer were used to evaluate the prediction model. The algorithm took 19 s to predict the results of 308 test images and the accuracy of the test results of the WLI and NBI esophageal cancer was 88 and 91%, respectively, when using the spectral data. Compared with RGB images, the accuracy of the WLI was 83% and the NBI was 86%. In this study, the accuracy of the WLI and NBI was increased by 5%, confirming that the prediction accuracy of the HSI detection method is significantly improved.

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.

Esophageal squamous‐cell neoplasia (ESCN) is a common second primary neoplasia found in patients with head‐and‐neck squamous‐cell carcinoma (HNSCC). This study sought to identify the risk factors for synchronous ESCN and how they influence survival in HNSCC patient. Eight hundred and fifteen incident HNSCC patients were prospectively recruited for endoscopy screening for ESCN using white‐light imaging, narrow‐band imaging, Lugol chromoendoscopy, and pathological confirmation. Associated lifestyle and clinicopathological data were collected. The interquartile follow‐up period cutoffs were 11.3, 20.5 and 34.9 months. 124 patients (15.2%) were diagnosed as having synchronous ESCN (66 low‐grade dysplasia, 29 high‐grade dysplasia, and 29 esophageal squamous‐cell carcinoma). Consumption of alcohol, but not betel nut or cigarette, was significantly associated with the presence of synchronous ESCN (adjusted odds ratio [aOR] = 7.1 and 10.9 for former and current drinkers, respectively). There was an interaction between cumulative dose of alcohol consumption and alcohol flushing response on the development of ESCN. High‐dose drinkers with flush response were 16.9 times more likely to have esophageal high‐grade dysplasia/SCC than non‐drinkers. Compared with oral cavity cancer patients, those with hypopharyngeal, laryngeal and oropharyngeal cancer were 6.8, 4.6 and 2.8 times more likely to have esophageal high‐grade dysplasia/SCC. HNSCC patients with synchronous ESCN had lower overall survival than those without (p < 0.0001). In conclusion, surveillance of ESCN is strongly recommended for the high‐risk subpopulation of HNSCC patients, especially drinkers who have a flush response to alcohol, and those with distant metastasis of index cancer and cancers in hypopharynx, oropharynx and larynx.

Diagnosis of early esophageal neoplasia, including dysplasia and superficial cancer, is a great challenge for endoscopists. Recently, the application of artificial intelligence (AI) using deep learning in the endoscopic field has made significant advancements in diagnosing gastrointestinal cancers. In the present study, we constructed a single-shot multibox detector using a convolutional neural network for diagnosing different histological grades of esophageal neoplasms and evaluated the diagnostic accuracy of this computer-aided system. A total of 936 endoscopic images were used as training images, and these images included 498 white-light imaging (WLI) and 438 narrow-band imaging (NBI) images. The esophageal neoplasms were divided into three classifications: squamous low-grade dysplasia, squamous high-grade dysplasia, and squamous cell carcinoma, based on pathological diagnosis. This AI system analyzed 264 test images in 10 s, and the sensitivity, specificity, and diagnostic accuracy of this system in detecting esophageal neoplasms were 96.2%, 70.4%, and 90.9%, respectively. The accuracy of this AI system in differentiating the histological grade of esophageal neoplasms was 92%. Our system showed better accuracy in diagnosing NBI (95%) than WLI (89%) images. Our results showed the great potential of AI systems in identifying esophageal neoplasms as well as differentiating histological grades.

Predicting complex disease risks on the basis of individual genomic profiles is an advancing field in human genetics^1,2. However, most genetic studies have focused on populations of European ancestry, creating a global imbalance in precision medicine and underscoring the need for genomic research in non-European groups^3,4. The Taiwan Precision Medicine Initiative recruited more than half a million Taiwanese residents, providing a large dataset of genetic profiles and electronic medical record data for people with Han Chinese ancestry. Using extensive phenotypic data, we conducted comprehensive genomic analyses across the medical phenome with individuals genetically similar to Han Chinese reference populations. These analyses identified population-specific genetic risk variants and new findings for various complex traits. We developed polygenic risk scores, demonstrating strong predictive performance for conditions such as cardiometabolic diseases, autoimmune disorders, cancers and infectious diseases. We observed consistent findings in an independent dataset, Taiwan Biobank, and among people of East Asian ancestry in the UK Biobank and the All of Us Project. The identified genetic risks accounted for up to 10.3% of the overall health variation in the Taiwan Precision Medicine Initiative cohort. Our approach of characterizing the phenome-wide genomic landscape, developing population-specific risk-prediction models, assessing their performance and identifying the genetic effect on health serves as a model for similar studies in other diverse study populations.

This study aimed to identify noninvasive protein markers capable of detecting the presence and prognosis of esophageal squamous-cell carcinoma (ESCC). Analyzing microarray expression data collected from 17-pair ESCC specimens, we identified one protein, matrix metalloproteinase-1 (MMP1), as a possibly useful marker. Plasma MMP1 was then measured by enzyme-linked immunosorbent assay (ELISA) in 210 ESCC patients and 197 healthy controls. ESCC patients had higher mean levels of MMP1 than controls (8.7 ± 7.5 vs. 6.7 ± 4.9 ng/mL, p < 0.0001). Using the highest quartile level (9.67 ng/mL) as cut-off, we found a 9.0-fold risk of ESCC in those with higher plasma MMP1 after adjusting for covariates (95% confidence interval = 2.2, 36.0). Heavy smokers and heavy drinkers with higher plasma MMP1 had 61.4- and 31.0 times the risk, respectively, than non-users with lower MMP1. In the survival analysis, compared to those with MMP1 ≤ 9.67 ng/mL, ESCC patients with MMP1 > 9.67 ng/mL had a 48% increase in the risk of ESCC death (adjusted hazard ratio = 1.48; 95% CI = 1.04–2.10). In conclusion, plasma MMP1 may serve as a noninvasive marker of detecting the presence and predicting the survival of ESCC.