BACKGROUND: Pre-operative combined modality therapy (CMT) is used in locally advanced rectal cancer. Its use affects the clinicopathological staging based on the resected specimen. Assessment of the tumour response in the resected specimen may provide prognostic information. This study was undertaken to determine the histological response to pre-operative chemoradiation and to assess the interobserver reliability of a newly developed tumour response grading system for rectal cancer. METHODS: Pre-operative biopsy specimens and the resected specimens of 21 patients with low rectal cancer were assessed. The patients underwent pre-operative CMT consisting of radiotherapy (45 Gy) with 5-FU either as a continuous infusion or as a bolus intravenous infusion with leucovorin. After four to six weeks tumour response was assessed by comparing pre-operative transrectal ultrasound (TRUS) findings (uT1-4, uN0-1) with postoperative histopathological assessment (pT1-4, pN0-1) using UICC TNM characteristics. Tumour response was defined as a decrease in T status. The histological response to CMT was based on the tumour regression grade (TRG) and ranged from fibrosis extending through the rectal wall with no residual cancer (TRG 1), to no evidence of tumour response (TRG 5). Inter-observer reliability was assessed using weighted and unweighted kappa statistics. RESULTS: Local downstaging was demonstrated in 11/21 (52%) of patients. Three of 21 patients had a TRG 1 response. Thirteen of 21 (62%) patients had TRG 1-3 responses to CMT. There was no significant correlation between local downstaging and TRG. The interobserver correlation coefficient for assessment of TRG was 0.88 (unweighted kappa). CONCLUSIONS: Local downstaging by pre-operative CMT can be demonstrated if pre-operative TRUS staging is compared to standard pathology staging in patients with rectal cancer. Local downstaging is not directly related to histologic response as assessed by TRG. Inter-observer reporting of tumour regression grade (TRG) is reliable.
The purpose of this pilot study is to determine whether a deep convolutional neural network can be trained with limited image data to detect high-grade small bowel obstruction patterns on supine abdominal radiographs. Grayscale images from 3663 clinical supine abdominal radiographs were categorized into obstructive and non-obstructive categories independently by three abdominal radiologists, and the majority classification was used as ground truth; 74 images were found to be consistent with small bowel obstruction. Images were rescaled and randomized, with 2210 images constituting the training set (39 with small bowel obstruction) and 1453 images constituting the test set (35 with small bowel obstruction). Weight parameters for the final classification layer of the Inception v3 convolutional neural network, previously trained on the 2014 Large Scale Visual Recognition Challenge dataset, were retrained on the training set. After training, the neural network achieved an AUC of 0.84 on the test set (95% CI 0.78-0.89). At the maximum Youden index (sensitivity + specificity-1), the sensitivity of the system for small bowel obstruction is 83.8%, with a specificity of 68.1%. The results demonstrate that transfer learning with convolutional neural networks, even with limited training data, may be used to train a detector for high-grade small bowel obstruction gas patterns on supine radiographs.
Objective: To evaluate the incidence and natural history of radiologic incisional hernia (IH) in patients who underwent robotic partial or radical nephrectomy. Materials and Methods: We retrospectively reviewed the records of patients who underwent robotic partial or radical nephrectomy for kidney tumor in our institution between January 2011 and April 2017. All pre-and postoperative imagings were re-reviewed for detection of IH per Tonouchi classification. Patients who developed hernia were followed up and classified into stable or progressive group. Clinical findings and radiologic features of these patients are reported. Results: A total of 247 patients (169 partial and 78 radical nephrectomies) were included in the study. The incidence of radiologic IH was 27.53%, graded as early-onset (35.3%), late-onset (51.5%), and bowel/fat containing (13.2%). Median time to radiologic IH was 1.7 years. During the follow-up of 68 patients who developed hernia, 33 (48.5%) had progressive and 8 (11.7%) developed clinical hernia. Median time to progression was 1.5 years. On multivariable analysis, adjuvant therapy was an independent predictor for radiologic hernia development (HR 3.23). Pathologic T stage ‡2 and history of open abdominal surgery were also significantly associated with hernia progression (HR 3.93 and 3.47, respectively). Conclusions: Radiologic IH after robotic partial or radical nephrectomy is common. Progression rate is as high as 50% with median time to progression of 1.5 years. Adjuvant therapy is an independent predictor for IH development, whereas higher stage and history of open abdominal surgery are associated with IH progression.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.