Deep learning analysis for differential diagnosis and risk classification of gastrointestinal tumors.

OBJECTIVES: Recently, artificial intelligence (AI) has been applied to clinical diagnosis. Although AI has already been developed for gastrointestinal (GI) tract endoscopy, few studies have applied AI to endoscopic ultrasound (EUS) images. In this study, we used a computer-assisted diagnosis (CAD) system with deep learning analysis of EUS images (EUS-CAD) and assessed its ability to differentiate GI stromal tumors (GISTs) from other mesenchymal tumors and their risk classification performance. MATERIALS AND METHODS: A total of 101 pathologically confirmed cases of subepithelial lesions (SELs) arising from the muscularis propria layer, including 69 GISTs, 17 leiomyomas and 15 schwannomas, were examined. A total of 3283 EUS images were used for training and five-fold-cross-validation, and 827 images were independently tested for diagnosing GISTs. For the risk classification of 69 GISTs, including very-low-, low-, intermediate- and high-risk GISTs, 2,784 EUS images were used for training and three-fold-cross-validation. RESULTS: For the differential diagnostic performance of GIST among all SELs, the accuracy, sensitivity, specificity and area under the receiver operating characteristic (ROC) curve were 80.4%, 82.9%, 75.3% and 0.865, respectively, whereas those for intermediate- and high-risk GISTs were 71.8%, 70.2%, 72.0% and 0.771, respectively. CONCLUSIONS: The EUS-CAD system showed a good diagnostic yield in differentiating GISTs from other mesenchymal tumors and successfully demonstrated the GIST risk classification feasibility. This system can determine whether treatment is necessary based on EUS imaging alone without the need for additional invasive examinations.

Effectiveness of endoscopic ultrasound-guided tissue acquisition with stereomicroscopic on-site evaluation for preoperative diagnosis of resectable or borderline resectable pancreatic cancer: a prospective study.

Background/Aims: To validate endoscopic ultrasound-guided tissue acquisition (EUS-TA) used in conjunction with stereomicroscopic on-site evaluation (SOSE) as a preoperative diagnostic tool for resectable pancreatic cancer (R-PC) and borderline resectable PC (BR-PC). Methods: Seventy-eight consecutive patients who underwent EUS-TA for suspected R-PC or BR-PC were enrolled. The primary endpoint was the sensitivity of EUS-TA together with SOSE based on the stereomicroscopically visible white core (SVWC) cutoff value. One or two sites were punctured by using a 22-gauge biopsy needle for EUS-TA, based on the SOSE findings. Results: We collected 99 specimens from 56 and 22 patients with R-PC and BR-PC, respectively. Based on the SOSE results, we performed 57 procedures with one puncture. The SVWC cutoff values were met in 73.7% and 73.1% of all specimens and in those obtained during the first puncture, respectively. The final diagnoses were malignant and benign tumors in 76 and two patients, respectively. The overall sensitivity, specificity, and accuracy of EUS-TA for the 78 lesions were 90.8%, 100%, and 91.0%, respectively. The sensitivity for malignant diagnosis based on the SVWC cutoff value were 89.5% and 90.4% for the first puncture and all specimens, respectively. Conclusions: The sensitivity of EUS-TA in conjunction with SOSE for malignancy diagnosis in patients with suspected R-PC or BR-PC was 90.4%.

Endoscopic Ultrasound-Guided Tissue Acquisition Using Fork-Tip Needle for Subepithelial Lesions: A Single-Center Validation Study.

BACKGROUND: The cutoff value for stereomicroscopic on-site evaluation (SOSE) in endoscopic ultrasound-guided tissue acquisition (EUS-TA) has high diagnostic sensitivity when a Franseen needle is employed for upper gastrointestinal subepithelial lesions (SELs) (stereomicroscopically visible white core [SVWC] ≥ 4 mm). AIM: We aimed to determine whether high diagnostic sensitivity could be obtained when EUS-TA was performed using a Fork-tip needle. METHODS: Twenty-one patients were prospectively registered. Patients underwent EUS-TA using a Fork-tip needle for upper gastrointestinal SELs at Kitasato University Hospital between January and November 2022. Punctures were made twice using the needle, and SOSE was conducted for each specimen. Blood and physical examination were performed to assess adverse events. Pathological diagnosis was made using hematoxylin and eosin-stained sections and immunohistochemical staining. Statistical comparisons were completed using Fisher's exact tests. RESULTS: The diagnostic rate of EUS-TA was 100% (21/21 cases). The final diagnosis was gastrointestinal stromal tumor in 17 (81.0%) and leiomyoma in 4 (19.0%) patients. SOSE was conducted on all 42 punctures, and the tissue sampling rate was 100% (42/42 punctures). Specimens with SVWC ≥ 4 mm were collected in 97.6% punctures (41/42 punctures) and the diagnostic sensitivity for these specimens was 100% (41/41 punctures), which is significantly higher (p < 0.0238) compared to the absence of cutoff value (diagnostic sensitivity of 0%). No EUS-TA-related adverse events occurred. CONCLUSIONS: EUS-TA combined with SOSE for upper gastrointestinal SEL using a fork-tip needle had a high diagnostic rate, and the cutoff value of SVWC ≥ 4 mm had high diagnostic sensitivity.

Efficacy of endoscopic ultrasound-guided tissue acquisition using stereo-microscopic on-site evaluation for possible comprehensive genome profile in patients with advanced pancreatic cancer.

BACKGROUND AND AIM: Stereomicroscopic on-site evaluation (SOSE) is a rapid evaluation method for endoscopic ultrasound-guided tissue acquisition (EUS-TA) with a high diagnostic sensitivity when the stereomicroscopically visible white core (SVWC) cut-off value (≥ 11 mm) is met. We prospectively examined the association between SVWCs and the adequacy of tissue specimens, assuming subsequent comprehensive genome profiling (CGP). METHODS: This study included 66 consecutive patients with suspected unresectable pancreatic cancer who underwent EUS-TA. The primary endpoint was the frequency of combined samples with ≥ 20% tumor cell content that met over twice the SVWC (T-SVWC) cut-off value, achieved through multiple punctures. The secondary endpoints were the number of punctures, the percentage of SVWC cut-off values, adverse events, the positive diagnosis rate, and the tissue section area. RESULTS: The median number of EUS-TA punctures for suspected unresectable pancreatic cancer was 3 (range, 3-4); SVWC and T-SVWC cut-off values were obtained in 171/206 specimens and 65/66 patients, respectively. There were no EUS-TA-related adverse events. The positive diagnosis rate of EUS-TA was 95.5%. Among the 63 patients meeting the T-SVWC cut-off value in pathological diagnoses, the median tumor cell content was 40% (range, 5-80%), with 57 patients having tumor cell content ≥ 20%. The median tissue section area was 15 (range, 3-40) mm2. CONCLUSIONS: When performing EUS-TA for unresectable pancreatic cancer with the intention of subsequent CGP, obtaining a high tumor cell content (≥ 20%) by assessing the T-SVWC cut-off value via SOSE may serve as a novel indicator for on-site estimation of CGP suitability for EUS-TA specimens.

Balloon enteroscopy-assisted endoscopic retrograde cholangiopancreatography for asymptomatic common bile duct stones on surgically altered anatomy: A high risk factor for post-endoscopic retrograde cholangiopancreatography pancreatitis.

BACKGROUND: Post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis (PEP) is more common in patients with asymptomatic common bile duct stones (CBDSs) and normal anatomy than in those with symptomatic CBDS; however, studies on the effect of surgically altered anatomy are lacking. We aimed to investigate whether asymptomatic CBDS in balloon enteroscopy-assisted ERCP cases for surgically altered anatomy also has a high incidence of PEP and analyze the risk factors for PEP. METHODS: We retrospectively analyzed 108 consecutive patients who underwent initial ERCP for CBDS with surgically altered anatomies and with naive papilla at Kitasato University Hospital from April 2015 to December 2022. RESULTS: Study participants were as follows: 92 (85%) patients with symptomatic CBDS and 16 (15%) patients asymptomatic CBDS. The overall bile duct cannulation success rate was 89.8%, with PEP occurring in 7.4% of patients (symptomatic CBDS: 3.3%, asymptomatic CBDS: 31.3%). PEP incidence was significantly higher for asymptomatic CBDS (p = .0017). Multivariate analysis identified asymptomatic CBDS and precut sphincterotomy as significant risk factors for PEP. CONCLUSIONS: Asymptomatic CBDS may be a risk factor for PEP onset in balloon enteroscopy-assisted ERCP with surgically altered anatomy. Therefore, the procedure should be performed after obtaining sufficient informed consent and adequate preparation.

Pancreatic Schwannoma Diagnosed by Endoscopic Ultrasound-guided Fine-needle Aspiration.

A schwannoma is a tumor originating from Schwann cells. It is occasionally observed in the abdominal viscera in the form of a submucosal tumor derived from the gastric or duodenal muscularis propria. To date, only a few studies have reported on pancreatic schwannomas. Furthermore, very few patients are preoperatively diagnosed with pancreatic schwannoma because of the lack of established imaging characteristics distinguishing this type of schwannoma from other conditions. We herein report the first English publication of pancreatic schwannoma in which surgery was avoided because a pathological diagnosis was made solely on the basis of endoscopic ultrasound-guided fine-needle aspiration findings.