Artificial intelligence-assisted esophageal cancer management: Now and future

Esophageal cancer (EC) is the seventh most common cancer and the sixth cause of cancer death worldwide. Histologically, esophageal squamous cell carcinoma (ESCC) and esophageal adenocarcinoma (EAC) account for up to 90% and 20% of all ECs, respectively. Clinical symptoms such as dysphagia, odynophagia, and bolus impaction occur late in the natural history of the disease, and the diagnosis is often delayed. The prognosis of ESCC and EAC is poor in advanced stages, being survival rates less than 20% at five years. However, when the diagnosis is achieved early, curative treatment is possible, and survival exceeds 80%. For these reasons, mass screening strategies for EC are highly desirable, and several options are currently under investigation. Blood biomarkers offer an inexpensive, non-invasive screening strategy for cancers, and novel technologies have allowed the identification of candidate markers for EC. The esophagus is easily accessible via endoscopy, and endoscopic imaging represents the gold standard for cancer surveillance. However, lesion recognition during endoscopic procedures is hampered by interobserver variability. To fill this gap, artificial intelligence (AI) has recently been explored and provided encouraging results. In this review, we provide a summary of currently available options to achieve early diagnosis of EC, focusing on blood biomarkers, advanced endoscopy, and AI.

Download Full-text

Modern Staging and Utility of PET Imaging in Esophageal Cancer Management

Journal of the National Comprehensive Cancer Network ◽

10.6004/jnccn.2008.0064 ◽

2008 ◽

Vol 6 (9) ◽

pp. 862-869 ◽

Cited By ~ 2

Author(s):

Kwang-Yu Chang ◽

Jang-Yang Chang ◽

Joseph Chao ◽

Yun Yen

Keyword(s):

Esophageal Cancer ◽

Neoadjuvant Therapy ◽

Imaging System ◽

False Negative ◽

Lymph Node Involvement ◽

Cancer Management ◽

Positron Emission ◽

Node Involvement ◽

Positive Results ◽

Locoregional Lymph Node

Esophageal cancer is the eighth most common cancer worldwide, and one of the most fatal diseases despite modern medical treatment. Because correct staging and surveillance of neoadjuvant therapy for esophageal cancer is mandatory for further treatment planning, choosing a modern imaging system is important. The development of 18F-fluorodeoxyglucose positron emission tomography (18F-FDG-PET) has provided alternate means of tumor detection distinct from more conventional methods. This modality has extraordinary performance in detecting locoregional lymph node involvement and distant metastatic disease, and has been introduced as a powerful tool in many guidelines. However, some factors still lead to false-negative or -positive results, raising questions of its accuracy. This article discusses the clinical efficacy of PET in staging and surveillance of neoadjuvant therapy in esophageal cancer, comparing its accuracy with conventional imaging modalities.

Download Full-text

Research and application of artificial intelligence based on electronic health record data from patients with cancer: a systematic review (Preprint)

10.2196/preprints.33799 ◽

2021 ◽

Author(s):

Xinyu Yang ◽

Dongmei Mu ◽

Hao Peng ◽

Hua Li ◽

Ying Wang ◽

...

Keyword(s):

Artificial Intelligence ◽

Electronic Health Records ◽

Ensemble Methods ◽

Future Research ◽

Electronic Health Record Data ◽

Health Records ◽

Patients With Cancer ◽

Cancer Management ◽

Treatment And Prevention ◽

Electronic Health

BACKGROUND With the accumulation of electronic health records data and the development of artificial intelligence, patients with cancer urgently need new evidence of more personalized clinical and demographic characteristics and more sophisticated treatment and prevention strategies. However, no research has systematically analyzed the application and significance of electronic health records and artificial intelligence in cancer care. OBJECTIVE In this study, we reviewed the literature on the application of AI based on EHR data from patients with cancer, hoping to provide reference for subsequent researchers, and help accelerate the application of EHR data and AI technology in the field of cancer, so as to help patients get more scientific and accurate treatment. METHODS Three databases were systematically searched to retrieve potentially relevant articles published from January 2009 to October 2020. A combination of terms related to "electronic health records", "artificial intelligence" and "cancer" was used to search for these publications. RESULTS Of the 1034 articles considered, 148 met the inclusion criteria. The review has shown that ensemble methods and deep learning were on the rise. It presented the representative literatures on the subfield of cancer diagnosis, treatment and care. In addition, the vast majority of studies in this area were based on private institutional databases, resulting in poor portability of the proposed methodology process. CONCLUSIONS The use of new methods and electronic health records data sharing and fusion were recommended for future research. With the help of specialists, artificial intelligence and the mining of massive electronic medical records could provide great opportunities for improving cancer management.

Download Full-text

Potential impact of preoperative EUS on esophageal cancer management and cost

Gastrointestinal Endoscopy ◽

10.1016/s0016-5107(02)70044-8 ◽

2002 ◽

Vol 56 (3) ◽

pp. 391-396 ◽

Cited By ~ 24

Author(s):

Douglas A. Shumaker ◽

Patricia de Garmo ◽

Douglas O. Faigel

Keyword(s):

Esophageal Cancer ◽

Cancer Management ◽

Potential Impact

Download Full-text

The Role of Artificial Intelligence (AI) in Cancer Treatment in Low-Middle Income Countries: The Watson for Oncology (WfO) Experience (Preprint)

10.2196/preprints.31461 ◽

2021 ◽

Author(s):

Angela Rui ◽

Srinivas Emani ◽

Hermano Alexandre Lima Rocha ◽

Rubina F. Rizvi ◽

Sergio Ferreira Juaçaba ◽

...

Keyword(s):

Artificial Intelligence ◽

Cancer Treatment ◽

Clinical Decision Support Systems ◽

Clinical Decision ◽

Cultural Barriers ◽

Successful Implementation ◽

Middle Income ◽

Middle Income Countries ◽

Cancer Management ◽

Low Middle Income Countries

UNSTRUCTURED As technology continues to improve, healthcare systems have the opportunity to utilize a variety of innovative tools for decision making that extend beyond traditional clinical decision support systems (CDSSs). The feasibility and efficacy integrating artificial intelligence (AI) systems into medical practice has shown variable success, especially in resource-poor areas. In this paper, we cover the existing challenges surrounding cancer treatment in low-middle income countries (LMICs). By focusing on the implementation of an AI-based CDSS for oncology, we aim to demonstrate how AI can be both beneficial and challenging for cancer management globally. Additionally, we summarize current physician perspectives from China, India, Brazil, Thailand, and Mexico in regard to their experiences and recommendations for improving the system. By doing so, we hope to highlight the need for additional research on user experience and unique cultural barriers for the successful implementation of AI in LMICs.

Download Full-text

The Use of Neural Networks in Identifying Risk Factors for Lymph Node Metastasis and Recommending Management of T1b Esophageal Cancer

The American Surgeon ◽

10.1177/000313481207800238 ◽

2012 ◽

Vol 78 (2) ◽

pp. 195-206 ◽

Cited By ~ 3

Author(s):

George Sgourakis ◽

Ines Gockel ◽

Orestis Lyros ◽

Sophocles Lanitis ◽

Georgia Dedemadi ◽

...

Keyword(s):

Neural Networks ◽

Esophageal Cancer ◽

Lymph Node ◽

Rank Order ◽

Lymph Node Status ◽

Histologic Type ◽

Level Of Evidence ◽

Cancer Management ◽

Number Of Patients ◽

Histologic Differentiation

The objective of this study was to establish a prediction model of lymph node status in T1b esophageal carcinoma and define the best squamous and adenocarcinoma predictors. The literature lacks a satisfactory level of evidence of T1b esophageal cancer management. We performed an analysis pooling the effects of outcomes of 2098 patients enrolled into 37 retrospective studies using “neural networks” as data mining techniques. The percentages for lymph node, lymphatic (L1), and vascular (V1) invasion in Sm1 esophageal cancers were 24, 46, and 20 per cent, respectively. The same parameters apply to Sm2 with 34, 63, and 38 per cent as opposed to Sm3 with 51, 69, and 47 per cent. The respective number of patients with well, moderate, and poor histologic differentiation totaled 267, 752, and 582. The rank order of the predictors of lymph node positivity was, respectively: Grade III, (L1), (V1), Sm3 invasion, Sm2 invasion, and Sm1 invasion. Histologic-type squamous and adenocarcinoma (ADC/SCC) was not included in the model. The best predictors for SCC lymph node positivity were sm3 invasion and (V1). As concerns ADC, the most important predictor was (L1). Submucosal esophageal cancer should be managed with surgical resection. However, this is subject to the histologic type and presence of specific predictors that could well alter the perspective of multimodality management.

Download Full-text