Does a learning curve exist in endorectal two-dimensional ultrasound accuracy?

Convolutional neural network deep learning model accurately detects rectal cancer in endoanal ultrasounds

BACKGROUND

Imaging is vital for assessing rectal cancer, with endoanal ultrasound (EAUS) being highly accurate in large tertiary medical centers. However, EAUS accuracy drops outside such settings, possibly due to varied examiner experience and fewer examinations. This underscores the need for an AI-based system to enhance accuracy in non-specialized centers. This study aimed to develop and validate deep learning (DL) models to differentiate rectal cancer in standard EAUS images.

METHODS

A transfer learning approach with fine-tuned DL architectures was employed, utilizing a dataset of 294 images. The performance of DL models was assessed through a tenfold cross-validation.

RESULTS

The DL diagnostics model exhibited a sensitivity and accuracy of 0.78 each. In the identification phase, the automatic diagnostic platform achieved an area under the curve performance of 0.85 for diagnosing rectal cancer.

CONCLUSIONS

This research demonstrates the potential of DL models in enhancing rectal cancer detection during EAUS, especially in settings with lower examiner experience. The achieved sensitivity and accuracy suggest the viability of incorporating AI support for improved diagnostic outcomes in non-specialized medical centers.

Intra- and Interobserver Variability of Shear Wave Elastography in Rectal Cancer

Simple Summary

For a diagnostic tool to be important, there must be a high level of agreement between different observers. If not, the reliability of the measurement can not be considered valid. Previous studies have evaluated the reproducibility of tissue stiffness measurements of rectal tumors, but only on previously obtained images. Therefore, we aimed to investigate the issue in a clinical setting and found a high level of agreement between observers.

Abstract

Background: Endorectal ultrasound (ERUS) is an important tool when evaluating complex rectal adenomas and rectal cancer, and the accuracy is improved by adding elastography measurements. A high interobserver agreement is imperative in clinical practice. Therefore, the aim of this study was to evaluate interobserver agreement assessed on real-time images. Additionally, we investigated the intra- and interobserver agreement between experienced and inexperienced observers. Materials and methods: We prospectively included patients referred to an ERUS at the Department of Radiology with a complex rectal polyp or suspected rectal malignancy. Two operators independently scanned each patient in turn. Furthermore, four observers assessed previously obtained images using three different methods for placing the region of interest (ROI). Three months later, the four observers reassessed the images to assess intraobserver variability. Results: A total of 19 patients were included for live assessment. Agreement of tumor classification was substantial for T stage (kappa: 0.86) and fair for N stage (kappa: 0.73), with an absolute agreement for T and N stages of 84% and 89%, respectively. Agreement of SWE was good for Emean (ICC 0.94, 95% CI 0.86–0.98) and fair for Emax (ICC 0.85, 95% CI 0.66–0.94). Intra- and interobserver agreement between inexperienced and experienced observers showed good to excellent agreement with all ROI methods. Conclusion: Interobserver agreement is high in SWE when performed in a clinical setting. We found the best agreement using the mean value of several ROIs. Intra- and interobserver agreement was high regardless of operator experience.

Can Ultrasound Elastography Discriminate between Rectal Adenoma and Cancer? A Systematic Review

BACKGROUND

Rectal cancer is a common malignancy. Since the introduction of bowel-screening programs, the number of patients with advanced adenomas and early rectal cancer has increased. Despite improved diagnostics, the discrimination between rectal adenomas and early rectal cancer (i.e., pT1-T2) remains challenging. The purpose of this systematic review was to evaluate the diagnostic performance of endorectal ultrasound (ERUS) elastography in discriminating rectal adenomas from cancer.

METHOD

Using PRISMA guidelines, a systematic search was performed on PubMed, Embase, and MEDLINE databases. Studies evaluating the primary staging of rectal adenomas and cancer using ERUS elastography were included.

RESULTS

Six studies were identified; three evaluated the discrimination between adenomas and cancer; two evaluated adenomas and early rectal cancer (i.e., pT1-T2); one evaluated performance on different T categories. All studies reported increased diagnostic accuracy of ERUS elastography compared to ERUS. Sensitivity, specificity and accuracy ranged 0.93-1.00, 0.83-1.00 and 0.91-1.00, respectively, when discriminating adenomas from cancer. In the differentiation between adenomas and early rectal cancer, the sensitivity, specificity and accuracy were 0.82-1.00, 0.86-1.00 and 0.84-1.00, respectively.

CONCLUSION

Elastography increases the accuracy of ERUS and may provide valuable information on malignant transformation of rectal lesions.

EFSUMB Recommendations for Gastrointestinal Ultrasound Part 3: Endorectal, Endoanal and Perineal Ultrasound

This article represents part 3 of the EFSUMB Recommendations and Guidelines for Gastrointestinal Ultrasound (GIUS). It provides an overview of the examination techniques recommended by experts in the field of endorectal/endoanal ultrasound (ERUS/EAUS), as well as perineal ultrasound (PNUS). The most important indications are rectal tumors and inflammatory diseases like fistula and abscesses in patients with or without inflammatory bowel disease (IBD). PNUS sometimes is more flexible and convenient compared to ERUS. However, the technique of ERUS is quite well established, especially for the staging of rectal cancer. EAUS also gained ground in the evaluation of perianal diseases like fistulas, abscesses and incontinence. For the staging of perirectal tumors, the use of PNUS in addition to conventional ERUS could be recommended. For the staging of anal carcinomas, PNUS can be a good option because of the higher resolution. Both ERUS and PNUS are considered excellent guidance methods for invasive interventions, such as the drainage of fluids or targeted biopsy of tissue lesions. For abscess detection and evaluation, contrast-enhanced ultrasound (CEUS) also helps in therapy planning.

Measures of trainee performance in advanced endoscopy: A systematic review

INTRODUCTION

The diversity, technical skills required, and risk inherent to advanced endoscopy techniques all contribute to complex training curricula and steep learning curves. Since trainees develop endoscopy skills at different rates, there has been a shift towards competency-based training and certification. Validated endoscopy performance measures for trainees are, therefore, necessary. The aim of this systematic review was to describe and critically assess the existing evidence regarding measures of performance for trainees in advanced endoscopy.

METHODS

A systematic review of the literature from January 1980 to January 2016 was carried out using the MEDLINE, EMBASE, CENTRAL, and ISI Web of knowledge databases. MeSH terms related to 'advanced endoscopy' and 'performance' were applied to a highly sensitive search strategy. The main outcomes were face, content, and construct validity, as well as reliability.

RESULTS

The literature search yielded 1,662 studies and 77 met the inclusion criteria after abstract and full-text review (endoscopic retrograde cholangiopancreatography (ERCP)=23, endoscopic ultrasound (EUS)=30, colonoscopic polypectomy (CP)=11, balloon-assisted enteroscopy (BAE)=7, luminal stenting=3, radiofrequency ablation (RFA)=2, and endoscopic muscosal resection (EMR)=1). Good validity and reliability were found for measurement tools of overall performance in ERCP, EUS and CP, with applications for both patient-based and simulator training models. A number of specific technical skills were also shown to be valid measures of performance. These include: selective biliary cannulation, sphincterotomy, biliary stent placement, stone extraction and procedure time for ERCP; pancreatic solid mass T-staging, EUS-guided fine needle aspiration (EUS-FNA) procedure time, number of EUS-FNA passes and puncture precision for EUS; procedure time and en bloc resection rate for CP; retrograde fluoroscopy time for BAE; and mean number of endoscopy sessions required to achieve complete eradication of intestinal metaplasia (CIEM) for RFA. The evidence for EMR and luminal stenting is of insufficient quality to make recommendations.

CONCLUSIONS

We have identified multiple valid and readily available performance measures for advanced endoscopy trainees for ERCP, EUS, CP, BAE and RFA procedures. These tools should be considered in advanced endoscopy training programs wishing to move away from apprenticeship-based training and towards competency-based learning with the help of patient-based and simulator tools.

Learning curve of endorectal ultrasonography in preoperative staging of rectal carcinoma

Accurate preoperative staging of rectal carcinoma is essential for optimal treatment. This study was designed to evaluate the accuracy and learning curve of endorectal ultrasonography (ERUS) in the preoperative staging of rectal carcinoma. We retrospectively analyzed the records of patients with rectal carcinoma who underwent preoperative ERUS followed by curative surgery at the Shanxi Province Tumor Hospital between January, 2007 and March, 2010. The patients were divided into three groups, namely A, B and C, depending on whether the examination was performed between January and December, 2007, between January and December, 2008 or between January, 2009 and March, 2010, respectively. Five physicians with no prior experience in ERUS performed the examinations. We compared the ERUS staging with the pathological findings using the tumor-node-metastasis (TNM) classification. The accuracy of ERUS in T and N staging after each additional consecutive 20 patients was calculated for physicians D, E and F. A total of 319 patients underwent ERUS prior to surgery. There were 38 patients in group A, 135 in group B and 146 in group C. Two of the five physicians performed only 47 of the 319 examinations, whereas the remaining 272 patients were examined by physicians D (n=162), E (n=64) and F (n=46). The overall accuracy in assessing the extent of rectal wall invasion (T) was 67%, with 16% of the cases overstaged and 17% understaged and the accuracy in assessing nodal involvement (N) was 66%, with 11% of the cases overstaged and 23% understaged. The total T and N staging accuracy of physicians D, E and F was 75 and 72%; 59 and 59%; and 50 and 52%, respectively. For physicians D, E and F, the accuracy of T and N staging after each additional 20 patients was calculated and the curve of the accuracy reached a plateau after physician D completed 80 cases. Therefore, ERUS is a valuable tool for assessing the depth of tumor invasion and it appears that after ~80 cases a physician may be considered able to apply it efficiently.

Imaging assessment of early rectal cancer

Early rectal cancer (ERC) is defined as invasive adenocarcinoma spreading into, but not beyond, the submucosa or muscularis propria-that is a Dukes'A: T1N0 or T2N0 tumour in the tumour node metastasis (TNM) classification (Taylor et al. 2008). Among these tumours it is suggested that the most superficial T1 tumours least likely to metastasize to local lymph nodes than adenocarcinoma invading deeper where the rich lymphatic and venous plexuses within the submucosa provide a mechanism for tumour spread beyond the rectum. Currently, only about 10 % of patients presenting symptomatically with rectal cancer are diagnosed with early disease; however, up to 30 % of screen detected cancers are being identified as Dukes'A. Thus, the overall detection of early stage tumours is likely to increase following greater implementation in screening programs. The goal of this invited review is to provide recommendations based on the consensus discussion on the information from preoperative imaging that is of relevance for clinical decision-making for patients with early rectal cancer.