Deep Learning Empowers Endoscopic Detection and Polyps Classification: A Multiple-Hospital Study.

The present study aimed to develop an AI-based system for the detection and classification of polyps using colonoscopy images. A total of about 256,220 colonoscopy images from 5000 colorectal cancer patients were collected and processed. We used the CNN model for polyp detection and the EfficientNet-b0 model for polyp classification. Data were partitioned into training, validation and testing sets, with a 70%, 15% and 15% ratio, respectively. After the model was trained/validated/tested, to evaluate its performance rigorously, we conducted a further external validation using both prospective (n = 150) and retrospective (n = 385) approaches for data collection from 3 hospitals. The deep learning model performance with the testing set reached a state-of-the-art sensitivity and specificity of 0.9709 (95% CI: 0.9646-0.9757) and 0.9701 (95% CI: 0.9663-0.9749), respectively, for polyp detection. The polyp classification model attained an AUC of 0.9989 (95% CI: 0.9954-1.00). The external validation from 3 hospital results achieved 0.9516 (95% CI: 0.9295-0.9670) with the lesion-based sensitivity and a frame-based specificity of 0.9720 (95% CI: 0.9713-0.9726) for polyp detection. The model achieved an AUC of 0.9521 (95% CI: 0.9308-0.9734) for polyp classification. The high-performance, deep-learning-based system could be used in clinical practice to facilitate rapid, efficient and reliable decisions by physicians and endoscopists.

Differences in gene mutations according to gender among patients with colorectal cancer.

BACKGROUND: The incidence, site distribution, and mortality rates of patients with colorectal cancer differ according to gender. We investigated gene mutations in colorectal patients and wanted to examine gender-specific differences. METHODS: A total of 1505 patients who underwent surgical intervention for colorectal cancer were recruited from March 2000 to January 2010 at Taipei Veterans' General Hospital and investigated for gene mutations in K-ras, N-ras, H-ras, BRAF, loss of 18q, APC, p53, SMAD4, TGF-ß, PIK3CA, PTEN, FBXW7, AKT1, and MSI. RESULTS: There were significant differences between male and female patients in terms of tumor location (p < 0.0001) and pathological stage (p = 0.011). The female patients had significantly more gene mutations in BRAF (6.4 vs. 3.3%, OR 1.985, p = 0.006), TGF-ß (4.7 vs. 2.5%, OR 1.887, p = 0.027), and revealed a MSI-high status (14.0 vs. 8.3%, OR 1.800, p = 0.001) than male patients. Male patients had significantly more gene mutations in N-ras (5.1 vs. 2.3%, OR 2.227, p = 0.012); however, the significance was maintained only for mutations in BRAF (OR 2.104, p = 0.038), MSI-high status (OR 2.003 p = 0.001), and N-ras (OR 3.000, p = 0.010) after the groups were divided by tumor site. CONCLUSION: Gene mutations in BRAF, MSI-high status, and N-ras differ according to gender among patients with colorectal cancer.

Adjuvant FOLFOX treatment for stage III colon cancer: how many cycles are enough?

PURPOSE: Adjuvant FOLFOX (5-fluorouracil and oxaliplatin) chemotherapy benefits stage III colon cancer patients. However, it still results in side effects and increased cost. Reducing cycles had been thought to decrease these problems. This retrospective study aimed to find the appropriate number of treatment cycles that are sufficient for treating these patients. PATIENTS AND METHODS: A total of 213 stage III colon cancer patients receiving adjuvant FOLFOX therapy were retrospectively recruited. Demographic data were collected for analysis. Survival analyses were performed between all cases of patients receiving above and below a certain cycle number. If a significant difference was reached at that cycle number, multivariate Cox Regression was performed with those factors resulting in p < 0.2 to assess the independent prognostic factors. RESULTS: The 5-year overall survival rate of patients was 77.9 %, and the 3-year disease-free survival was 76.7 %. For overall survival, a significant benefit was noted for treatment of at least 8 cycles, for disease-free survival, significant differences were apparent from patient data of those who underwent from 7 to 12 treatment cycles. Multivariate survival analysis of that patient data at cycle 8 for overall survival and cycle 7 for disease free survival revealed cycle number as the only independent prognostic factor (p = 0.04, 0.048). CONCLUSION: Cycle number of adjuvant FOLFOX is a significant prognostic factor for stage III colon cancer patients. At least 8 cycles are needed to have an overall survival benefit, and 7 to disease-free survival.