FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384820938> ?p ?o ?g. }

• W4384820938 abstract "Endoscopic ultrasound (EUS) is used not only for diagnosing pancreatobiliary diseases but also for diagnosing subepithelial lesions (SELs) and the depth of gastrointestinal cancer, which are usually difficult to diagnose using endoscopy alone. The diagnosis of SELs is based on the layer of origin and the echo pattern; however, differential diagnosis of hypoechoic lesions originating from the fourth layer is difficult, since echo patterns of gastrointestinal stromal tumors (GISTs) with malignant potential is quite similar to those of other benign tumors, including leiomyoma and neurilemmoma.1, 2 Conventional tissue sampling using biopsy forceps cannot be applied to make a pathological diagnosis of SELs that are covered with normal mucosa.3 Endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) is generally used to obtain tissue samples from SELs.1-3 However, the diagnostic yield of EUS-FNA for SELs is lower than that for pancreatic solid tumors, since it is difficult to obtain sufficient specimens, especially for immunohistochemical staining, to make differential diagnoses of spindle cell tumors.2-4 Recently, unroofing biopsy techniques have been reported as an alternative to EUS-FNA;3 however, they are still not commonly used. Because of these diagnostic difficulties, studies are reporting the use of EUS image-based scoring systems or computer-aided diagnosis, such as artificial intelligence (AI)-aided diagnosis.5, 6 Several similar studies are expected to be conducted in the future. Conventional surgery, laparoscopic surgery, and laparoscopy and endoscopy cooperative surgery are performed for the treatment of SELs. Recently, minimally invasive treatments such as peroral endoscopic tumor resection, submucosal tunneling endoscopic resection (STER), and endoscopic full-thickness resection (EFTR), which are based on endoscopic submucosal dissection technology,7, 8 have been developed. These minimally invasive treatments may become mainstream treatments for SELs, depending on the lesion size. However, endoscopic resection (ER)-related problems such as hemorrhage during or after ER and the management of full-thickness defects, remain a problem. In this issue of Digestive Endoscopy, Chen et al.9 conducted a multicenter study to develop and evaluate an application system for determining the management of SELs. First, they retrospectively evaluated 837 cases with SELs in the esophagus, stomach, duodenum, and colorectum, which were obtained from eight tertiary hospitals in Shandong Province, China, from January 1, 2013 to April 30, 2020, as a derivation cohort. Several factors pertaining to patients' and lesions' characteristics and ER-related outcomes, including the clinical course, were evaluated and used to develop the application named “SELERP.” Subsequently, for a validation cohort, they evaluated 200 cases prospectively recruited from the same eight hospitals from May 1, 2020, to January 31, 2021. In this validation study, the sensitivities and specificities of the five diagnostic models (GIST, leiomyoma, neuroendocrine neoplasm, ectopic pancreas, and lipoma) were 68.3–95.7% and 64.1–83.3%, respectively. As a meta-analysis of previous EUS image-based AI-assisted diagnostic studies showed sensitivities and specificities of 0.93 (95% confidence interval [CI] 0.93–0.97) and 0.88 (95% CI 0.71–0.96), respectively,6 the diagnostic results of this study seemed to be somewhat inferior to those of previous studies. However, previous studies were mainly focused on the distinction between GIST and non-GIST, and the results of this study were considered satisfactory for evaluating the EUS findings alone. Although the prediction models derived for ER-related outcomes showed convincing discrimination and calibration, further examination will be conducted by considering more factors such as the treatment skill of endoscopists, the more segmented site of lesions, and the endoscopic treatment methods to improve this prediction model. This study had some limitations, such as the limited area of the derivation and validation studies and the need for further external validation at several other facilities. Furthermore, the endoscopists evaluated the endoscopic and EUS features themselves. However, by simply entering a few variables such as age, sex, lesion location, and endoscopic and EUS findings, SELERP could predict the diagnosis of SELs, use of conventional clips, procedure time, and other ER-related outcomes. This application can also be used on mobile phones, allowing easy access and seamless assessment in near real-time, which will be useful in clinical practice. In particular, it may be helpful in facilities that have started performing STER or EFTR. With the continued development of computer-aided diagnosis and assessment, it will become impossible to distinguish between human and artificial assessment, just as Dick felt.10 The day may come when human judgment may be deemed unnecessary. However, even with such an excellent application, so far only human beings can perform EUS, determine the need for treatment based on the diagnosis, and actually perform endoscopic treatments. Human training, and not just computer training, remains important. Author declares no conflict of interest for this article. Author S.Y. is supported by The National Cancer Center Research and Development Fund 2023-A-15." @default.
• W4384820938 created "2023-07-21" @default.
• W4384820938 creator A5028671078 @default.
• W4384820938 date "2023-07-20" @default.
• W4384820938 modified "2023-09-25" @default.
• W4384820938 title "Endoscopic ultrasound‐based application for determining the management of subepithelial lesions: Do androids dream of endoscopic ultrasound?" @default.
• W4384820938 cites W1939733260 @default.
• W4384820938 cites W2594013873 @default.
• W4384820938 cites W2820845148 @default.
• W4384820938 cites W2949004475 @default.
• W4384820938 cites W2954853016 @default.
• W4384820938 cites W4291000080 @default.
• W4384820938 cites W4294127381 @default.
• W4384820938 cites W4313539393 @default.
• W4384820938 cites W4365503187 @default.
• W4384820938 doi "https://doi.org/10.1111/den.14629" @default.
• W4384820938 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37469303" @default.
• W4384820938 hasPublicationYear "2023" @default.
• W4384820938 type Work @default.
• W4384820938 citedByCount "0" @default.
• W4384820938 crossrefType "journal-article" @default.
• W4384820938 hasAuthorship W4384820938A5028671078 @default.
• W4384820938 hasBestOaLocation W43848209381 @default.
• W4384820938 hasConcept C126838900 @default.
• W4384820938 hasConcept C142724271 @default.
• W4384820938 hasConcept C16930146 @default.
• W4384820938 hasConcept C2775934546 @default.
• W4384820938 hasConcept C2776170712 @default.
• W4384820938 hasConcept C2777007597 @default.
• W4384820938 hasConcept C2777795800 @default.
• W4384820938 hasConcept C2778451229 @default.
• W4384820938 hasConcept C2779066768 @default.
• W4384820938 hasConcept C2780390042 @default.
• W4384820938 hasConcept C2780801072 @default.
• W4384820938 hasConcept C71924100 @default.
• W4384820938 hasConceptScore W4384820938C126838900 @default.
• W4384820938 hasConceptScore W4384820938C142724271 @default.
• W4384820938 hasConceptScore W4384820938C16930146 @default.
• W4384820938 hasConceptScore W4384820938C2775934546 @default.
• W4384820938 hasConceptScore W4384820938C2776170712 @default.
• W4384820938 hasConceptScore W4384820938C2777007597 @default.
• W4384820938 hasConceptScore W4384820938C2777795800 @default.
• W4384820938 hasConceptScore W4384820938C2778451229 @default.
• W4384820938 hasConceptScore W4384820938C2779066768 @default.
• W4384820938 hasConceptScore W4384820938C2780390042 @default.
• W4384820938 hasConceptScore W4384820938C2780801072 @default.
• W4384820938 hasConceptScore W4384820938C71924100 @default.
• W4384820938 hasLocation W43848209381 @default.
• W4384820938 hasLocation W43848209382 @default.
• W4384820938 hasOpenAccess W4384820938 @default.
• W4384820938 hasPrimaryLocation W43848209381 @default.
• W4384820938 hasRelatedWork W2049331396 @default.
• W4384820938 hasRelatedWork W2111423964 @default.
• W4384820938 hasRelatedWork W2126658009 @default.
• W4384820938 hasRelatedWork W2321741795 @default.
• W4384820938 hasRelatedWork W2379854920 @default.
• W4384820938 hasRelatedWork W2508175788 @default.
• W4384820938 hasRelatedWork W2527386574 @default.
• W4384820938 hasRelatedWork W2612260388 @default.
• W4384820938 hasRelatedWork W4220723225 @default.
• W4384820938 hasRelatedWork W4384296733 @default.
• W4384820938 isParatext "false" @default.
• W4384820938 isRetracted "false" @default.
• W4384820938 workType "article" @default.