Multi-Task Model for Esophageal Lesion Analysis Using Endoscopic Images: Classification with Image Retrieval and Segmentation with Attention

The automatic analysis of endoscopic images to assist endoscopists in accurately identifying the types and locations of esophageal lesions remains a challenge. In this paper, we propose a novel multi-task deep learning model for automatic diagnosis, which does not simply replace the role of endoscopists in decision making, because endoscopists are expected to correct the false results predicted by the diagnosis system if more supporting information is provided. In order to help endoscopists improve the diagnosis accuracy in identifying the types of lesions, an image retrieval module is added in the classification task to provide an additional confidence level of the predicted types of esophageal lesions. In addition, a mutual attention module is added in the segmentation task to improve its performance in determining the locations of esophageal lesions. The proposed model is evaluated and compared with other deep learning models using a dataset of 1003 endoscopic images, including 290 esophageal cancer, 473 esophagitis, and 240 normal. The experimental results show the promising performance of our model with a high accuracy of 96.76% for the classification and a Dice coefficient of 82.47% for the segmentation. Consequently, the proposed multi-task deep learning model can be an effective tool to help endoscopists in judging esophageal lesions.

Spectrum of esophageal lesions- A study of endoscopic biopsies

Endoscopic biopsies is a very common outpatient procedure in gastroenterology unit. The present article is a study of endoscopic biopsies taken from the esophagus. The gastrointestinal tract cannot be visualized directly and endoscopy is a very important and easy tool for visualization of GI tract lesions directly and helping in the diagnosis in the early stage, there by helping in the overall well being of the patient. To emphasize the utility of endoscopic biopsies in the diagnosis of esophageal lesions. To correlate the endoscopic findings with pathological diagnosis. To study the various pathologies. Endoscopic biopsies taken from the esophagus were studied and analysed in the present study. The biopsies are taken by the gastroenterologist. The biopsies are fixed in 10%formalin.After fixation the biopsy specimen is processed and embedded in paraffin.4 to 5 microns thick sections were cut, stained with Haematoxylin & Eosin and studied. Out of the 118 cases of esophageal biopsies received at the Department of Pathology, Mediciti Institute of Medical Sciences from December 2016 to November, 2018. ,50 were non neoplastic lesions, 68 cases were neoplastic lesions, 50 were non neoplastic lesions, 68 cases were neoplastic lesions. Chronic non specific esophagitis was the commonest non neoplastic esophageal lesion. Squamous cell carcinoma was the commonest neoplastic lesion. The lower one third is the commonest site of pathology for esophagus. Males are mostly effected and predominant age of presentation is around 40-60 years for non neoplastic lesion and 50-70 years for neoplastic region.

Safety and Efficacy of High Power Shorter Duration Ablation Guided by Ablation Index or Lesion Size Index in Atrial Fibrillation Ablation: A Systematic Review and Meta-Analysis

Background. High power shorter duration (HPSD) ablation may lead to safe and rapid lesion formation. However, the optimal radio frequency power to achieve the desired ablation index (AI) or lesion size index (LSI) is insubstantial. This analysis aimed to appraise the clinical safety and efficacy of HPSD guided by AI or LSI (HPSD-AI or LSI) in patients with atrial fibrillation (AF). Methods. The Medline, PubMed, Embase, Web of Science, and the Cochrane Library databases from inception to November 2020 were searched for studies comparing HPSD-AI or LSI and low power longer duration (LPLD) ablation. Results. Seven trials with 1013 patients were included in the analysis. The analyses verified that HPSD-AI or LSI revealed benefits of first-pass pulmonary vein isolation (PVI) (RR: 1.28; 95% CI: 1.05–1.56, P = 0.01) and acute pulmonary vein reconnection (PVR) (RR: 0.65; 95% CI: 0.48–0.88, P = 0.005) compared with LPLD. HPSD-AI or LSI showed higher freedom from atrial tachyarrhythmia (AT) (RR = 1.32, 95% CI: 1.14–1.53, P = 0.0002) in the subgroup analysis of studies with PVI ± (with or without additional ablation beyond PVI). HPSD-AI or LSI could short procedural time (WMD: −22.81; 95% CI, −35.03 to −10.60, P = 0.0003), ablation time (WMD: −10.80; 95% CI: −13.14 to −8.46, P < .00001), and fluoroscopy time (WMD: −7.71; 95% CI: −13.71 to −1.71, P = 0.01). Major complications and esophageal lesion in HPSD-AI or LSI group were no more than LDLP group (RR: 0.58; 95% CI: 0.20–1.69, P = 0.32) and (RR: 0.84; 95% CI: 0.43–1.61, P = 0.59). Conclusions. HPSD-AI or LSI was efficient for treating AF with shorting procedural, ablation, and fluoroscopy time, higher first-pass PVI, and reducing acute PVR and may increase freedom from AT for patients with additional ablation beyond PVI compared with LPLD. Moreover, complications and esophageal lesion were low and no different between two groups.

Esophagitis Dissecans Superficialis (EDS) Secondary to Hair Dye Ingestion: Case Report and Literature Review

Esophagitis dissecans superficialis (EDS) is a rare and underdiagnosed esophageal lesion characterized by sloughing of the esophageal mucosa that has been associated with medications, various autoimmune disorders, and exposure to some chemical irritants. Anatomically, EDS is most commonly seen in the middle and distal thirds of the esophagus. When present, EDS is best treated by discontinuing the offending agent and initiating pharmacologic therapy with proton pump inhibitors. Steroids may also be effective if the etiology is autoimmune in nature. Our case highlights a 65-year-old female diagnosed with EDS after incidental ingestion of hair dye containing resorcinol and para-phenylenediamine (PPD).

Additional data on protection of the esophagus during catheter ablation of atrial fibrillation

The excellent review by Houmsse and Daoud of techniques and methods utilized to protect the esophagus from injury during atrial fibrillation (AF) ablation appropriately concludes that considering the ease of use, minimal side effects, and low costs associated with esophageal protection devices, compelling evidence exists for use of esophageal protection as routine care for AF ablation. Some additional data are available which would warrant inclusion in further consideration of this topic. Three recent studies have demonstrated the inability of LET monitoring to protect the esophagus, whereas meta-analysis of three studies of manual cooling using direct liquid instillation suggests that this approach significantly reduced high-grade lesion formation (OR of 0.39, 95% CI 0.17 to 0.89). Moreover, three studies using a commercially available cooling device FDA cleared for thermal regulation have shown reductions in esophageal lesion severity without degradation in ablation efficacy.

Further mechanistic evidence against luminal esophageal temperature monitoring?

We congratulate Kar et al. on their elegant study evaluating ex-vivo temperature profiles and the resulting thermal injury formation on the epiesophageal surface during radiofrequency (RF) ablation. In addition to being the first study to detail temperature profiles inclusive of the epiesophageal surface during RF ablation, we believe that the results add further concern to the use of temperature sensing technology in the quest to reduce esophageal injury. Three recent clinical trials have evaluated the efficacy of luminal esophageal temperature (LET) monitoring and found either no benefits, or signals of harm. On the other hand, two pilot RCTs suggest benefits of active cooling, and a large RCT, the IMPACT study, further confirms this benefit by finding an 83% reduction in esophageal lesion formation using an active cooling device. With no degradation in ablation efficacy, as well as a reduction/elimination of the need for fluoroscopy and reports of shortened procedure time with active cooling technology, the data of Kar et al., combined with growing clinical data, suggest that continued use of LET monitoring may be unjustified.

Aorto-esophageal fistula caused by fishbone ingestion: a case report on staged endovascular and endoscopic treatment

Background Aorto-esophageal fistula (AEF) caused by foreign bodies ingestion is a rare but devastating disorder. Thoracic endovascular aortic repair (TEVAR) has become a widely accepted intervention for treating aorto-esophageal fistulas. As for post-TEVAR esophageal defect, secondary esophagectomy has been the recommended choice for most of the AEFs, but there is no general consensus with regard to the need of secondary surgeries for patients in the absence of clear signs of reinfection or bleeding. We herein presented a case of an AEF caused by fishbone ingestion, after successful TEVAR, the esophageal lesion was closed endoscopically. Case presentation A 38-year-old male presented with esophageal fistula for 4 months. He was diagnosed with AEF because of Chiari's triad after fishbone ingestion 4 months ago. Emergency thoracic aortic stent implantation was done, and given broad spectrum antibiotics and blood transfusion. His symptoms were improved, and discharged with an esophageal fistula left to heal itself. Nevertheless, after 4 months, re-examination of esophago-gastro-duodenoscopy revealed that the diameter of the fistula was changed from 3 to 6 mm. He was then admitted to our hospital for esophageal fistula repair. Laboratory examinations and chest computed tomography showed no signs of active infection, and endoscopic closure of the fistula was achieved with 4 clips. After that, he was discharged and gradually returned to normal diet. Conclusion For AEFs in the absence of active infection with repaired aorta but persistent esophageal fistula, endoscopic closure by endoclips might be an effective treatment choice.