Artificial intelligence using deep learning analysis of endoscopic ultrasonography images for the differential diagnosis of pancreatic masses.

BACKGROUND : There are several types of pancreatic mass, so it is important to distinguish between them before treatment. Artificial intelligence (AI) is a mathematical technique that automates learning and recognition of data patterns. This study aimed to investigate the efficacy of our AI model using endoscopic ultrasonography (EUS) images of multiple types of pancreatic mass (pancreatic ductal adenocarcinoma [PDAC], pancreatic adenosquamous carcinoma [PASC], acinar cell carcinoma [ACC], metastatic pancreatic tumor [MPT], neuroendocrine carcinoma [NEC], neuroendocrine tumor [NET], solid pseudopapillary neoplasm [SPN], chronic pancreatitis, and autoimmune pancreatitis [AIP]). METHODS : Patients who underwent EUS were included in this retrospective study. The included patients were divided into training, validation, and test cohorts. Using these cohorts, an AI model that can distinguish pancreatic carcinomas from noncarcinomatous pancreatic lesions was developed using a deep-learning architecture and the diagnostic performance of the AI model was evaluated. RESULTS : 22 000 images were generated from 933 patients. The area under the curve, sensitivity, specificity, and accuracy (95 %CI) of the AI model for the diagnosis of pancreatic carcinomas in the test cohort were 0.90 (0.84-0.97), 0.94 (0.88-0.98), 0.82 (0.68-0.92), and 0.91 (0.85-0.95), respectively. The per-category sensitivities (95 %CI) of each disease were PDAC 0.96 (0.90-0.99), PASC 1.00 (0.05-1.00), ACC 1.00 (0.22-1.00), MPT 0.33 (0.01-0.91), NEC 1.00 (0.22-1.00), NET 0.93 (0.66-1.00), SPN 1.00 (0.22-1.00), chronic pancreatitis 0.78 (0.52-0.94), and AIP 0.73 (0.39-0.94). CONCLUSIONS : Our developed AI model can distinguish pancreatic carcinomas from noncarcinomatous pancreatic lesions, but external validation is needed.

Current status of artificial intelligence analysis for endoscopic ultrasonography.

Endoscopic ultrasonography (EUS) is an essential diagnostic tool for various types of pancreatic diseases such as pancreatic tumors and chronic pancreatitis; however, EUS imaging has low specificity for the diagnosis of pancreatic diseases. Artificial intelligence (AI) is a mathematical prediction technique that automates learning and recognizes patterns in data. This review describes the details and principles of AI and deep learning algorithms. The term AI does not have any definite definition; almost all AI systems fall under narrow AI, which can handle single or limited tasks. Deep learning is based on neural networks, which is a machine learning technique that is widely used in the medical field. Deep learning involves three phases: data collection and annotation, building the deep learning architecture, and training and ability validation. For medical image diagnosis, image classification, object detection, and semantic segmentation are performed. In EUS, AI is used for detecting anatomical features, differential pancreatic tumors, and cysts. For this, conventional machine learning architectures are used, and deep learning architecture has been used in only two reports. Although the diagnostic abilities in these reports were about 85-95%, these were exploratory research and very few reports have included substantial evidence. AI is increasingly being used for medical image diagnosis due to its high performance and will soon become an essential technique for medical diagnosis.

Clinical utility of endoscopic ultrasound-guided tissue acquisition for comprehensive genomic profiling of pancreatic cancer.

BACKGROUND/AIMS: Endoscopic ultrasound-guided tissue acquisition (EUS-TA) is essential for the diagnosis of pancreatic cancer. The feasibility of comprehensive genomic profiling (CGP) using samples obtained by EUS-TA has been under recent discussion. This study aimed to evaluate the utility of EUS-TA for CGP in a clinical setting. METHODS: CGP was attempted in 178 samples obtained from 151 consecutive patients with pancreatic cancer at the Aichi Cancer Center between October 2019 and September 2021. We evaluated the adequacy of the samples for CGP and determined the factors associated with the adequacy of the samples obtained by EUS-TA retrospectively. RESULTS: The overall adequacy for CGP was 65.2% (116/178), which was significantly different among the four sampling methods (EUS-TA vs. surgical specimen vs. percutaneous biopsy vs. duodenal biopsy, 56.0% [61/109] vs. 80.4% [41/51] vs. 76.5% [13/17] vs. 100.0% [1/1], respectively; p=0.022). In a univariate analysis, needle gauge/type was associated with adequacy (22 G fine-needle aspiration vs. 22 G fine-needle biopsy [FNB] vs. 19 G-FNB, 33.3% (5/15) vs. 53.5% (23/43) vs. 72.5% (29/40); p=0.022). The sample adequacy of 19 G-FNB for CGP was 72.5% (29/40), and there was no significant difference between 19 G-FNB and surgical specimens (p=0.375). CONCLUSION: To obtain adequate samples for CGP with EUS-TA, 19 G-FNB was shown to be the best in clinical practice. However, 19 G-FNB was not still sufficient, so further efforts are required to improve adequacy for CGP.

Utility of forward-view endoscopic ultrasound in fine-needle aspiration in patients with a surgically altered upper gastrointestinal anatomy.

BACKGROUND/AIMS: Endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA) using oblique-view EUS in patients with a surgically altered anatomy (SAA) of the upper gastrointestinal tract is limited because of difficult scope insertion due to the disturbed anatomy. This study aimed to investigate the efficiency of forward-view (FV)-EUS in performing FNA in patients with a SAA. METHODS: We retrospectively investigated 32 patients with a SAA of the upper gastrointestinal tract who visited Aichi Cancer Center Hospital in Nagoya, Japan, between January 2014 and December 2020. We performed-upper gastrointestinal EUS-FNA using FV-EUS combined with fluoroscopic imaging to confirm tumor recurrence or to make a decision before chemotherapy or after a failure of diagnosis by radiology. RESULTS: We successfully performed EUS-FNA in all studied patients (100% technical success), with the specificity, sensitivity, and accuracy of 100%, 87.5%, and 87.8%, respectively, with no complications. CONCLUSION: EUS-FNA using FV-EUS combined with fluoroscopic imaging is an effective and safe technique for tissue acquisition in patients with a SAA.

Endoscopic ultrasound-guided portal vein coiling: troubleshooting interventional endoscopic ultrasonography.

Endoscopic ultrasound (EUS)-guided hepaticogastrostomy (HGS) is widely performed not only as an alternative to transpapillary biliary drainage, but also as primary drainage for malignant biliary obstruction. For anatomical reasons, this technique carries an unavoidable risk of mispuncturing intrahepatic vessels. We report a technique for troubleshooting EUS-guided portal vein coiling to prevent bleeding from the intrahepatic portal vein after mispuncture during interventional EUS. EUS-HGS was planned for a 59-year-old male patient with unresectable pancreatic cancer. The dilated bile duct (lumen diameter, 2.8 mm) was punctured with a 19-gauge needle, and a guidewire was inserted. After bougie dilation, the guidewire was found to be inside the intrahepatic portal vein. Embolizing coils were placed to prevent bleeding. Embolization coils were successfully inserted under stabilization of the catheter using a double-lumen cannula with a guidewire. Following these procedures, the patient was asymptomatic. Computed tomography performed the next day revealed no complications.

B2 puncture with forward-viewing EUS simplifies EUS-guided hepaticogastrostomy (with video).

Background and Objectives: EUS-guided hepaticogastrostomy (EUS-HGS) is in widespread use; however, there are few dedicated devices. The B2 route is technically easier than the B3 route for guidewire insertion, dilation, and stenting but if performed with conventional oblique-viewing (OV) EUS, B2 puncture can cause transesophageal puncture and severe adverse events. The aim of this study was to assess the efficacy of forward-viewing (FV) EUS, which we have developed to improve safety for B2 puncture in EUS-HGS (B2-EUS-HGS). Patients and Methods: This single-center retrospective study included 61 consecutive patients who underwent B2-EUS-HGS with FV between February 2020 and March 2021 at Aichi Cancer Center, Japan. The patients were prospectively enrolled, and clinical data were retrospectively collected for these 61 cases. Results: The overall technical success rate of EUS-HGS was 98.3% (60/61). The rate of EUS-HGS with FV was 95.0% (58/61) after three cases converted to OV, and that of B2-EUS-HGS with FV was 88.5% (54/61). The early adverse event rate was 6.5% (4/61). There were no instances of transesophageal puncture. Median procedure time was 24 min (range, 8-70), and no patient required cautery dilation. Conclusions: B2-EUS-HGS can be performed safely using FV, without transesophageal puncture, and supportability of the device is improved as FV is coaxial with the guidewire. FV was efficacious in B2-EUS-HGS, which shows promise for clinical application in the future.

Outcomes of Endoscopic Ultrasound-Guided Biliary Drainage in Patients Undergoing Antithrombotic Therapy.

BACKGROUND/AIMS: The Japan Gastroenterological Endoscopy Society (JGES) has published guidelines for gastroenterological endoscopy in patients undergoing antithrombotic treatment. These guidelines classify endoscopic ultrasound-guided biliary drainage (EUS-BD) as a high-risk procedure. Nevertheless, the bleeding risk of EUS-BD in patients undergoing antithrombotic therapy is uncertain. Therefore, this study aimed to assess the bleeding risk in patients undergoing antithrombotic therapy. METHODS: This single-center retrospective study included 220 consecutive patients who underwent EUS-BD between January 2013 and December 2018. We managed the withdrawal and continuation of antithrombotic agents according to the JGES guidelines. We compared the bleeding event rates among patients who received and those who did not receive antithrombotic agents. RESULTS: A total of 18 patients (8.1%) received antithrombotic agents and 202 patients (91.8%) did not. Three patients experienced bleeding events, with an overall bleeding event rate of 1.3% (3/220): one patient was in the antithrombotic group (5.5%) and two patients were in the non-antithrombotic group (0.9%) (p=0.10). All cases were moderate. The sole thromboembolic event (0.4%) was a cerebral infarction in a patient in the non-antithrombotic group. CONCLUSION: The rate of EUS-BD-related bleeding events was low. Even in patients receiving antithrombotic therapy, the bleeding event rates were not significantly different from those in patients not receiving antithrombotic therapy.

Pregnancy Is a Risk Factor for Secondary Focal Segmental Glomerulosclerosis in Women with a History of Very Low Birth Weight.

Low birth weight (LBW) has been known to increase the susceptibility to renal injury in adulthood. A 26-year-old woman developed proteinuria in early pregnancy; she had been born with very LBW. The clinical course was progressive, and an emergency Caesarean section was performed at 36 weeks due to acute kidney injury. A renal biopsy provided a diagnosis of post-adaptive focal segmental glomerulosclerosis. Increased demand for glomerular filtration during early pregnancy appeared to have initiated the renal injury. This report highlights the fact that pregnancy might be a risk factor for renal injury in women born with LBW.

Preeclampsia before 20 weeks of gestation: a case report and review of the literature.

The occurrence of preeclampsia before 20 weeks of gestation is rare and usually associated with trophoblastic diseases or antiphospholipid syndrome. Here, we report a case of preeclampsia before 20 weeks of gestation in the absence of the aforementioned disorders. A healthy 30-year-old nulliparous woman presented with new onset of hypertension and proteinuria at 18 weeks of gestation. Fetal ultrasound did not reveal any abnormalities. Empirical steroid treatment was initiated based on a tentative diagnosis of underlying renal disease. The clinical course of the disease was progressive despite steroid treatment and the fetus died in utero 8 days after the initiation of treatment. Following delivery, a renal biopsy was performed and provided a diagnosis of preeclampsia. All symptoms resolved postpartum. This report demonstrates that preeclampsia may occur before 20 weeks of gestation and should always be considered in the differential diagnosis of pregnant women with new onset of hypertension with proteinuria. Previous published cases are summarized briefly.