Artificial intelligence in endoscopic diagnosis of esophageal squamous cell carcinoma and precancerous lesions

Sep 23, 2026

Esophageal squamous cell carcinoma (ESCC) remains a major global health challenge requiring early detection and timely diagnosis to improve outcomes. Endoscopic examination is central, yet available endoscopic techniques can still miss or misdiagnose ESCC due to limitations such as subtle early features and operator variability. Artificial intelligence (AI)-assisted endoscopic diagnosis has shown progress, supporting more accurate characterization, margin delineation, invasion depth estimation, and microvascular subtype classification while guiding future advancements.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Abstract

Introduction

Endoscopic examination techniques and limitations

AI-assisted deep learning approaches

Deep learning-assisted detection of lesions

WLI modality

Review Article Artificial intelligence in endoscopic diagnosis of esophageal squamous cell carcinoma and precancerous lesions Nuoya Zhou1, Xianglei Yuan2, Wei Liu1, Qi Luo1, Ruide Liu1, Bing Hu1 1Department of Gastroenterology and Hepatology, West China Hospital, Sichuan University, Chengdu, Sichuan 610041, China; 2Digestive Endoscopy Medical Engineering Research Laboratory, West China Hospital, Med-X Center for Materials, Sichuan University, Chengdu, Sichuan 610041, China. Abstract Esophageal squamous cell carcinoma (ESCC) poses a significant global health challenge, necessitating early detection, timely diagnosis, and prompt treatment to improve patient outcomes. Endoscopic examination plays a pivotal role in this regard. However, despite the availability of various endoscopic techniques, certain limitations can result in missed or misdiagnosed ESCCs. Currently, artificial intelligence (AI)-assisted endoscopic diagnosis has made significant strides in addressing these limitations and improving the diagnosis of ESCC and precancerous lesions. In this review, we provide an overview of the current state of AI applications for endoscopic diagnosis of ESCC and precancerous lesions in aspects including lesion characterization, margin delineation, invasion depth estimation, and microvascular subtype classification. Furthermore, we offer insights into the future direction of this field, highlighting potential advancements that can lead to more accurate diagnoses and ultimately better prognoses for patients. Keywords: Artificial intelligence; Endoscopy; Esophageal squamous cell carcinoma; Precancerous lesions Introduction In recent years, there has been a significant expansion in esophageal cancer (EC) screening, driven by rapid advancements in endoscopic techniques and the widespread adoption of screening initiatives. However, EC remains a formidable global health challenge, ranking as the sixth most prevalent cancer globally and the third leading cause of cancer-related mortality, according to the current cancer statistics. [1] EC comprises two distinct histological subtypes: esophageal squamous cell carcinoma (ESCC) and esophageal adenocarcinoma. Despite a declining trend in ESCC incidence, it remains the predominant form of EC in Asia and certain parts of sub-Saharan Africa. [2] The prognosis of ESCC patients is closely related to the clinical stage at the time of diagnosis. Most patients are diagnosed at an advanced stage, resulting in an overall 5-year survival rate less than 30% even after treatments like surgery and chemotherapy. [3–5] Conversely, patients diagnosed with early-stage ESCC can achieve remarkable 5-year survival rates of up to 90% when following appropriate treatment regimens. [6] Therefore, early detection, timely diagnosis, and prompt treatment are crucial factors in improving the prognosis of patients with ESCC. | Access this article online | | | --- | --- | | Quick Response Code: | Website:
[www.cmj.org](www.cmj.org) | | | DOI:
10.1097/CM9.0000000000003490 | Endoscopic examination and treatment play pivotal roles in achieving early diagnosis and treatment of ESCC. Several endoscopic examination techniques are available, including white light imaging (WLI), chromoendoscopy (e.g., iodine staining), magnifying endoscopy (ME), electronic staining endoscopy such as narrow band imaging (NBI), and blue laser imaging (BLI) . The routine endoscopic esophageal examination procedure typically starts with the use of WLI or non-ME NBI/BLI (non-magnifying endoscopy with narrow band imaging or blue laser imaging) to identify suspicious lesions. [7] Subsequently, iodine staining or ME-NBI/BLI is employed to aid in characterizing these lesions. [7] Despite the array of available endoscopic tools, certain limitations persist, leading to the missed diagnoses of ESCC. These limitations include subtle microendoscopic features of early lesions, variations inendoscopists’ expertise, and the impact of v

FAQs

Why is early detection and timely diagnosis crucial for ESCC prognosis?

ESCC prognosis depends strongly on clinical stage at diagnosis. Many patients present at advanced stages with poor survival, while early-stage patients can achieve much higher 5-year survival with appropriate treatment.

What endoscopic techniques are commonly used to evaluate suspected esophageal lesions?

The review describes approaches including white light imaging (WLI), chromoendoscopy (e.g., iodine staining), magnifying endoscopy (ME), electronic staining modalities such as narrow band imaging (NBI), and blue laser imaging (BLI). Suspicious lesions are first identified, then further characterized using iodine staining or ME-NBI/BLI.

Which AI application areas are covered for improving endoscopic diagnosis of ESCC and precancerous lesions?

The review focuses on lesion characterization, margin delineation, invasion depth estimation, and microvascular subtype classification, aiming to reduce missed or misdiagnosed cases and improve diagnostic accuracy.

Sep 23, 2026

Uploaded by นรินทร์