Artificial intelligence for the real-time classification of intrapapillary capillary loop patterns in the endoscopic diagnosis of early oesophageal squamous cell carcinoma: A proof-of-concept study.

Background: Intrapapillary capillary loops (IPCLs) represent an endoscopically visible feature of early squamous cell neoplasia (ESCN) which correlate with invasion depth - an important factor in the success of curative endoscopic therapy. IPCLs visualised on magnification endoscopy with Narrow Band Imaging (ME-NBI) can be used to train convolutional neural networks (CNNs) to detect the presence and classify staging of ESCN lesions. Methods: A total of 7046 sequential high-definition ME-NBI images from 17 patients (10 ESCN, 7 normal) were used to train a CNN. IPCL patterns were classified by three expert endoscopists according to the Japanese Endoscopic Society classification. Normal IPCLs were defined as type A, abnormal as B1-3. Matched histology was obtained for all imaged areas. Results: This CNN differentiates abnormal from normal IPCL patterns with 93.7% accuracy (86.2% to 98.3%) and sensitivity and specificity for classifying abnormal IPCL patterns of 89.3% (78.1% to 100%) and 98% (92% to 99.7%), respectively. Our CNN operates in real time with diagnostic prediction times between 26.17 ms and 37.48 ms. Conclusion: Our novel and proof-of-concept application of computer-aided endoscopic diagnosis shows that a CNN can accurately classify IPCL patterns as normal or abnormal. This system could be used as an in vivo, real-time clinical decision support tool for endoscopists assessing and directing local therapy of ESCN.

Endoscopic management of Barrett's and early oesophageal neoplasia.

Barrett's oesophagus (BO) is the only known precursor to oesophageal adenocarcinoma (OAC). Dysplasia and intramucosal cancer arising in BO can safely be treated with endoscopic eradication therapy (EET) due to the low risk of subsequent lymph node metastasis. Treatment at an early stage is paramount due to the ongoing poor prognosis and outcomes of patients with advanced OAC. The mainstay of treatment is endoscopic resection of visible lesions for accurate staging followed by ablation therapy to all remaining columnar-lined epithelium, most commonly with radiofrequency ablation. Successful eradication of dysplasia can be achieved in >95% of patients with this EET combined approach.

Improvement over time in outcomes for patients undergoing endoscopic therapy for Barrett's oesophagus-related neoplasia: 6-year experience from the first 500 patients treated in the UK patient registry.

BACKGROUND: Barrett's oesophagus (BE) is a pre-malignant condition leading to oesophageal adenocarcinoma (OAC). Treatment of neoplasia at an early stage is desirable. Combined endoscopic mucosal resection (EMR) followed by radiofrequency ablation (RFA) is an alternative to surgery for patients with BE-related neoplasia. METHODS: We examined prospective data from the UK registry of patients undergoing RFA/EMR for BE-related neoplasia from 2008 to 2013. Before RFA, visible lesions were removed by EMR. Thereafter, patients had RFA 3-monthly until all BE was ablated or cancer developed (endpoints). End of treatment biopsies were recommended at around 12 months from first RFA treatment or when endpoints were reached. Outcomes for clearance of dysplasia (CR-D) and BE (CR-IM) at end of treatment were assessed over two time periods (2008-2010 and 2011-2013). Durability of successful treatment and progression to OAC were also evaluated. RESULTS: 508 patients have completed treatment. CR-D and CR-IM improved significantly between the former and later time periods, from 77% and 56% to 92% and 83%, respectively (p<0.0001). EMR for visible lesions prior to RFA increased from 48% to 60% (p=0.013). Rescue EMR after RFA decreased from 13% to 2% (p<0.0001). Progression to OAC at 12 months is not significantly different (3.6% vs 2.1%, p=0.51). CONCLUSIONS: Clinical outcomes for BE neoplasia have improved significantly over the past 6 years with improved lesion recognition and aggressive resection of visible lesions before RFA. Despite advances in technique, the rate of cancer progression remains 2-4% at 1 year in these high-risk patients. TRIAL REGISTRATION NUMBER: ISRCTN93069556.