Regular Mucosa Research Articles

Deep learning in CT colonography: differentiating premalignant from benign colorectal polyps

ObjectivesTo investigate the differentiation of premalignant from benign colorectal polyps detected by CT colonography using deep learning.MethodsIn this retrospective analysis of an average risk colorectal cancer screening sample, polyps of all size categories and morphologies were manually segmented on supine and prone CT colonography images and classified as premalignant (adenoma) or benign (hyperplastic polyp or regular mucosa) according to histopathology. Two deep learning models SEG and noSEG were trained on 3D CT colonography image subvolumes to predict polyp class, and model SEG was additionally trained with polyp segmentation masks. Diagnostic performance was validated in an independent external multicentre test sample. Predictions were analysed with the visualisation technique Grad-CAM++.ResultsThe training set consisted of 107 colorectal polyps in 63 patients (mean age: 63 ± 8 years, 40 men) comprising 169 polyp segmentations. The external test set included 77 polyps in 59 patients comprising 118 polyp segmentations. Model SEG achieved a ROC-AUC of 0.83 and 80% sensitivity at 69% specificity for differentiating premalignant from benign polyps. Model noSEG yielded a ROC-AUC of 0.75, 80% sensitivity at 44% specificity, and an average Grad-CAM++ heatmap score of ≥ 0.25 in 90% of polyp tissue.ConclusionsIn this proof-of-concept study, deep learning enabled the differentiation of premalignant from benign colorectal polyps detected with CT colonography and the visualisation of image regions important for predictions. The approach did not require polyp segmentation and thus has the potential to facilitate the identification of high-risk polyps as an automated second reader.Key Points• Non-invasive deep learning image analysis may differentiate premalignant from benign colorectal polyps found in CT colonography scans.• Deep learning autonomously learned to focus on polyp tissue for predictions without the need for prior polyp segmentation by experts.• Deep learning potentially improves the diagnostic accuracy of CT colonography in colorectal cancer screening by allowing for a more precise selection of patients who would benefit from endoscopic polypectomy, especially for patients with polyps of 6–9 mm size.

Machine Learning–based Differentiation of Benign and Premalignant Colorectal Polyps Detected with CT Colonography in an Asymptomatic Screening Population: A Proof-of-Concept Study

Background CT colonography does not enable definite differentiation between benign and premalignant colorectal polyps. Purpose To perform machine learning-based differentiation of benign and premalignant colorectal polyps detected with CT colonography in an average-risk asymptomatic colorectal cancer screening sample with external validation using radiomics. Materials and Methods In this secondary analysis of a prospective trial, colorectal polyps of all size categories and morphologies were manually segmented on CT colonographic images and were classified as benign (hyperplastic polyp or regular mucosa) or premalignant (adenoma) according to the histopathologic reference standard. Quantitative image features characterizing shape (n = 14), gray level histogram statistics (n = 18), and image texture (n = 68) were extracted from segmentations after applying 22 image filters, resulting in 1906 feature-filter combinations. Based on these features, a random forest classification algorithm was trained to predict the individual polyp character. Diagnostic performance was validated in an external test set. Results The random forest model was fitted using a training set consisting of 107 colorectal polyps in 63 patients (mean age, 63 years ± 8 [standard deviation]; 40 men) comprising 169 segmentations on CT colonographic images. The external test set included 77 polyps in 59 patients comprising 118 segmentations. Random forest analysis yielded an area under the receiver operating characteristic curve of 0.91 (95% CI: 0.85, 0.96), a sensitivity of 82% (65 of 79) (95% CI: 74%, 91%), and a specificity of 85% (33 of 39) (95% CI: 72%, 95%) in the external test set. In two subgroup analyses of the external test set, the area under the receiver operating characteristic curve was 0.87 in the size category of 6-9 mm and 0.90 in the size category of 10 mm or larger. The most important image feature for decision making (relative importance of 3.7%) was quantifying first-order gray level histogram statistics. Conclusion In this proof-of-concept study, machine learning-based image analysis enabled noninvasive differentiation of benign and premalignant colorectal polyps with CT colonography. © RSNA, 2021 Online supplemental material is available for this article.

A Careful Preoperative Airway Assessment

Clinical Image: A 45-year-old male subject with recent diagnosis of right colon cancer was referred to Department of Otorhinolaryngology for a preoperative airway assessment. The patient presented a long years history of intermittent hoarseness, retrosternal burning and rare episodes of laryngospasm, which however have never been further investigated. Videolaryngoscopy examination (Figure 1) showed a pinkish, smooth and hard mass in the supraglottis, covered by regular mucosa, narrowing anterior glottic region, which demonstrated the left vocal fold fixed in paramedial position. The subglottic space appeared well preserved. CT scan of the neck (Figures 2 and 3) revealed an hypodense mass originating from left aryepiglottic fold incontiguity with the left arytenoids and expanded from the posteromedial left side to superomedial side. The lesion presented a peripheral calcified stippled appearance and was surrounded by discreet soft -tissue masses, without enhancement of intravenous contrast. No enlarged lymph nodes were visible. The imaging findigs are indicative of a congenital laryngeal malformation.The airway space was judged sufficient for endotracheal intubation to perform abdominal surgery. No introperative or extubation complications occured. The patient will be introduced to a dedicated follow-up endoscopy to carry out biopsy and surgical treatment of the laryngeal lesion.

Up-regulation of Neutrophil Gelatinase-Associated Lipocalin in Colorectal Cancer Predicts Poor Patient Survival

Lipocalin-2 (Lcn-2) is expressed in human neutrophils and epithelial cells, particularly in the presence of inflammation or cancer. It was shown to be highly expressed in various human cancers. Increased protein levels were associated with decreased survival of patients with breast or gastric cancer. The main focus of this work was to analyze the implication of Lcn-2 up-regulation in the genesis of colon cancer. Expression of Lcn-2 was analyzed in colorectal carcinoma cell lines, paired colorectal carcinoma tissues, and regular mucosa by Western blot analysis. Lcn-2 immunohistochemical staining was performed in 192 colorectal carcinoma resection specimens and correlated with clinicopathologic parameters. Western blot analysis of colorectal carcinoma tissues demonstrated Lcn-2 overexpression in carcinomas as compared with regular mucosa. Immunohistochemical staining revealed Lcn-2 expression in 179 (93.2%) colorectal carcinoma tissues. Intense immunoreactivity was significantly correlated with metastasis (p = 0.042) and UICC stage (p = 0.027). Survival analysis according to the Kaplan-Meier method revealed a significant association between Lcn-2 overexpressing tumors and overall survival (p < 0.001) and disease-free survival (p < 0.001). Our data provide evidence that Lcn-2 expression is up-regulated with tumor progression and was found to be a predictor of overall survival.

Laryngeal Chondroma: A Rare Diagnosis in This Localization

Primary chondroid tumors of the larynx represent less than 1% of all laryngeal tumors. Most of them are chondromas and they often involve to the cricoid cartilage. A 31-year-old male patient applied to the oto-laryngology service with a history of dysphonia and dyspnea. Microlaryngoscopy revealed 2 cm sized, ill-defined, covered with regular mucosa, porous, and hard mass on posterior surface of crycoid cartilage in subglottic area. Following the excision of the lesion, histopathologic examination revealed as chondroma. Two years later, local recurrence was detected and the diagnosis was again chondroma. There was no complaint of the patient in last 3 and half years of follow-up. Chondroma should carefully be differentiated from chondrosarcoma and the patients should be followed for possible recurrences.

S2057 Inhibiting Bacterial Translocation Through M-Cells - Evaluation of Soluble Plantain Fibers As a Potential Prophylactic Therapy for Infective Diarrhea

with a diameter of 0.9mm was introduced through the instrumentation channel of the endoscope and positioned on neoplastic or non-neoplastic mucosa. Micro vessel density was calculated with a specially designed software. Results: The endoscopic-endomicroscopic examination with differentiation between regular mucosa and cancerous lesions was possible in all animals. Micro-vessel density was significantly higher in mice with cancers compared to regular mucosa (18308.5 ± 790.6μm2 vs. 15013.5 ± 1071.7μm2; p=0.003; t-test). Conclusion: pCLE with colonoscopy helps to detect and monitor carcinogenesis and angiogenesis in a novel cancer mouse model. These data might help to understand endoscopic changes during carcinogenesis but might also facilitate monitoring of anti-angiogenic therapy.

Effect of the width of excisional mucosa stoma ring on postoperative results of procedure for prolapse and hemorrhoids method

To explore the effect of the width of excisional mucosal stoma ring on postoperative results of procedure for prolapse and hemorrhoids (PPH) in the treatment of hemorrhoids. Four hundred and eighty hemorrhoid patients were treated with circular stapler. Whether excisional mucosa stoma ring was regular or not, and its width were recorded. After follow-up from one month to 24 months,the outcomes were compared. The mean age of the patients was (46.6+/- 5.1) years. Regular mucosa stoma ring was excited in 303 cases,while irregular ring in 177 cases. Hemorrhoids prolapse occurred in 18 cases of the patients with irregular mucosal ring excited, while only 8 cases of the patients with regular ring excited (P< 0.01). There was significant difference in the incidence of hemorrhoids prolapse between the different groups according to different width of excisional mucosal stoma ring (P< 0.01), and the patients with postoperative hemorrhoids prolapse was lest with the width of excisional mucosal stoma ring more than 3 cm. The width of excisional mucosal stoma ring is one of the effecting factors of surgical outcomes. It is effective to excite mucosal stoma ring more than 2 cm,and most effective to excite mucosal stoma ring more than 3 cm.

NO TITLE AVAILABLE

A 49yrs old, male patient presented progressive dysphagia during the last two years, initially to solids and, since two weeks, liquids he had lost 15 kilos. Esophagogram showed an intraluminal lesion with regular mucosa in the middle third of esophagus. The esophagoscopy exhibits regular mucosa up to the lesion, which did not alow passage. There was no evidence of linfonodal disease at CT scan. The hypothesis of esophageal leiomyoma led this patient a right posterior thoracotomy and enucleation of a 30 x 30 mm mass from the esophageal wall. With this case report we intend to review esophageal leiomyomas and their diagnosis and treatment.

Titanium as a biomaterial for ossicular replacement: results after implantation in the middle ear of the rabbit.

The middle ear poses unique challenges when finding suitable materials for ossicular reconstruction, primarily because of its link to the external environment via the eustachian tube and, hence, its greater exposure to infectious agents. In this study, the biocompatability of titanium was examined in the middle ear of rabbits by using light and scanning electron microscopy. Implants were placed as middle ear prostheses or as free implants. These were inspected at 28 days, 84 days, 168 days, 336 days and 504 days following implantation for mucosal coverage, percent epithelization and any sign of foreign-body reaction. After 28 days, the prostheses were covered by regular mucosa. Although a majority of the free implants took up to 336 days for complete epithelialization, some of the free implants were not epithelialized even at day 504. There were no inflammatory cells observed on the surface of the material, nor were unusual amounts of fibrous tissue seen. In addition, the titanium material exhibited an affinity toward bone. The results of this animal experiment indicate that titanium is a favorable material for ossicular replacement prostheses.

Epithelialization of titanium prostheses in the middle ear of the rabbit. Possible model of mucosa development

The epithelialization of middle ear implants is regarded as a sign of biocompatibility of the implant material. The mucosal coverage of titanium was studied in the middle ear of rabbits in light microscopy. Implants were used as middle ear prostheses or as a free implants. Studies were performed after 28, 84, 168, 336, and 504 days. The prostheses were seen to be covered by regular mucosa after 28 days. The free implants were not seen to be totally epithelialized even after 504 days. There were no inflammation cells on the surface of the material, nor were unusual amounts of fibrous tissue observed. The middle ear mucosa was initially thickened, but soon returned to normal values. The results of this animal experiment show that titanium is a suitable material for ossicular replacement.