Accurate Classification of Diminutive Colorectal Polyps Using Computer-Aided Analysis.

BACKGROUND & AIMS: Narrow-band imaging is an image-enhanced form of endoscopy used to observed microstructures and capillaries of the mucosal epithelium which allows for real-time prediction of histologic features of colorectal polyps. However, narrow-band imaging expertise is required to differentiate hyperplastic from neoplastic polyps with high levels of accuracy. We developed and tested a system of computer-aided diagnosis with a deep neural network (DNN-CAD) to analyze narrow-band images of diminutive colorectal polyps. METHODS: We collected 1476 images of neoplastic polyps and 681 images of hyperplastic polyps, obtained from the picture archiving and communications system database in a tertiary hospital in Taiwan. Histologic findings from the polyps were also collected and used as the reference standard. The images and data were used to train the DNN. A test set of images (96 hyperplastic and 188 neoplastic polyps, smaller than 5 mm), obtained from patients who underwent colonoscopies from March 2017 through August 2017, was then used to test the diagnostic ability of the DNN-CAD vs endoscopists (2 expert and 4 novice), who were asked to classify the images of the test set as neoplastic or hyperplastic. Their classifications were compared with findings from histologic analysis. The primary outcome measures were diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic time. The accuracy, sensitivity, specificity, PPV, NPV, and diagnostic time were compared among DNN-CAD, the novice endoscopists, and the expert endoscopists. The study was designed to detect a difference of 10% in accuracy by a 2-sided McNemar test. RESULTS: In the test set, the DNN-CAD identified neoplastic or hyperplastic polyps with 96.3% sensitivity, 78.1% specificity, a PPV of 89.6%, and a NPV of 91.5%. Fewer than half of the novice endoscopists classified polyps with a NPV of 90% (their NPVs ranged from 73.9% to 84.0%). DNN-CAD classified polyps as neoplastic or hyperplastic in 0.45 ± 0.07 seconds-shorter than the time required by experts (1.54 ± 1.30 seconds) and nonexperts (1.77 ± 1.37 seconds) (both P < .001). DNN-CAD classified polyps with perfect intra-observer agreement (kappa score of 1). There was a low level of intra-observer and inter-observer agreement in classification among endoscopists. CONCLUSIONS: We developed a system called DNN-CAD to identify neoplastic or hyperplastic colorectal polyps less than 5 mm. The system classified polyps with a PPV of 89.6%, and a NPV of 91.5%, and in a shorter time than endoscopists. This deep-learning model has potential for not only endoscopic image recognition but for other forms of medical image analysis, including sonography, computed tomography, and magnetic resonance images.

Adaptation of influenza A (H7N9) virus in primary human airway epithelial cells.

Influenza A (H7N9) is an emerging zoonotic pathogen with pandemic potential. To understand its adaptation capability, we examined the genetic changes and cellular responses following serial infections of A (H7N9) in primary human airway epithelial cells (hAECs). After 35 serial passages, six amino acid mutations were found, i.e. HA (R54G, T160A, Q226L, H3 numbering), NA (K289R, or K292R for N2 numbering), NP (V363V/I) and PB2 (L/R332R). The mutations in HA enabled A(H7N9) virus to bind with higher affinity (from 39.2% to 53.4%) to sialic acid α2,6-galactose (SAα2,6-Gal) linked receptors. A greater production of proinflammatory cytokines in hAECs was elicited at later passages together with earlier peaking at 24 hours post infection of IL-6, MIP-1α, and MCP-1 levels. Viral replication capacity in hAECs maintained at similar levels throughout the 35 passages. In conclusion, during the serial infections of hAECs by influenza A(H7N9) virus, enhanced binding of virion to cell receptors with subsequent stronger innate cell response were noted, but no enhancement of viral replication could be observed. This indicates the existence of possible evolutional hurdle for influenza A(H7N9) virus to transmit efficiently from human to human.

Rectal cancer with complete clinical response after neoadjuvant chemoradiotherapy, surgery, or "watch and wait".

PURPOSE: The purpose of this study was to compare the outcomes of patients treated with chemoradiotherapy with a complete clinical response followed by either a "watch and wait" strategy or a total mesorectal excision. METHODS: This was an observational retrospective study from a single institute. Patients with locally advanced rectal cancer following chemoradiotherapy with a complete clinical response from January 1, 2007 to December 31, 2014 were included. RESULTS: The study population consisted of 18 patients who opted for a "watch and wait" policy and 26 patients who underwent radical surgery after achieving a complete clinical response. Patients had no documented treatment complications under the watch and wait policy, while 13 patients who underwent radical surgery experienced significant morbidity. There were two local recurrences in the watch and wait group; both were treated with salvage resection and had no associated mortality. In the radical surgery group, 1 patient showed an incomplete pathologic response (ypT0N1), and the remaining 25 patients showed complete pathologic responses; 1 had a distant recurrence, which was managed non-operatively, and 2 patients died of unrelated causes. The 5-year overall survival rate and median disease-free survival time were 100% and 69.78 months in the watch and wait group and 92.30% and 89.04 months in the radical surgery group. CONCLUSIONS: A watch and wait policy avoids the morbidity associated with radical surgery and preserves oncologic outcomes in our retrospective study from a single institute. It could be considered a therapeutic option in patients with locally advanced rectal cancer following chemoradiotherapy with a complete clinical response.

Molecular epidemiology and phylodynamics of the human respiratory syncytial virus fusion protein in northern Taiwan.

BACKGROUND AND AIMS: The glycoprotein (G protein) and fusion protein (F protein) of respiratory syncytial virus (RSV) both show genetic variability, but few studies have examined the F protein gene. This study aimed to characterize the molecular epidemiology and phylodynamics of the F protein gene in clinical RSV strains isolated in northern Taiwan from 2000-2011. METHODS: RSV isolates from children presenting with acute respiratory symptoms between July 2000 and June 2011 were typed based on F protein gene sequences. Phylogeny construction and evaluation were performed using the neighbor-joining (NJ) and maximum likelihood (ML) methods. Phylodynamic patterns in RSV F protein genes were analyzed using the Bayesian Markov Chain Monte Carlo framework. Selection pressure on the F protein gene was detected using the Datamonkey website interface. RESULTS: From a total of 325 clinical RSV strains studied, phylogenetic analysis showed that 83 subgroup A strains (RSV-A) could be further divided into three clusters, whereas 58 subgroup B strains (RSV-B) had no significant clustering. Three amino acids were observed to differ between RSV-A and -B (positions 111, 113, and 114) in CTL HLA-B*57- and HLA-A*01-restricted epitopes. One positive selection site was observed in RSV-B, while none was observed in RSV-A. The evolution rate of the virus had very little change before 2000, then slowed down between 2000 and 2005, and evolved significantly faster after 2005. The dominant subtypes of RSV-A in each epidemic were replaced by different subtypes in the subsequent epidemic. CONCLUSIONS: Before 2004, RSV-A infections were involved in several small epidemics and only very limited numbers of strains evolved and re-emerged in subsequent years. After 2005, the circulating RSV-A strains were different from those of the previous years and continued evolving through 2010. Phylodynamic pattern showed the evolutionary divergence of RSV increased significantly in the recent 5 years in northern Taiwan.