Methods for improving colorectal cancer annotation efficiency for artificial intelligence-observer training

BACKGROUND Missing occult cancer lesions accounts for the most diagnostic errors in retrospective radiology reviews as early cancer can be small or subtle,making the lesions difficult to detect.Secondobserver is the most effective technique for reducing these events and can be economically implemented with the advent of artificial intelligence(AI).AIM To achieve appropriate AI model training,a large annotated dataset is necessary to train the AI models.Our goal in this research is to compare two methods for decreasing the annotation time to establish ground truth:Skip-slice annotation and AI-initiated annotation.METHODS We developed a 2D U-Net as an AI second observer for detecting colorectal cancer(CRC)and an ensemble of 5 differently initiated 2D U-Net for ensemble technique.Each model was trained with 51 cases of annotated CRC computed tomography of the abdomen and pelvis,tested with 7 cases,and validated with 20 cases from The Cancer Imaging Archive cases.The sensitivity,false positives per case,and estimated Dice coefficient were obtained for each method of training.We compared the two methods of annotations and the time reduction associated with the technique.The time differences were tested using Friedman’s two-way analysis of variance.RESULTS Sparse annotation significantly reduces the time for annotation particularly skipping 2 slices at a time(P<0.001).Reduction of up to 2/3 of the annotation does not reduce AI model sensitivity or false positives per case.Although initializing human annotation with AI reduces the annotation time,the reduction is minimal,even when using an ensemble AI to decrease false positives.CONCLUSION Our data support the sparse annotation technique as an efficient technique for reducing the time needed to establish the ground truth.

Electro-and photochemical studies of gold (Ⅲ) bromide towards a novel laser-based method of gold patterning

In this report, we demonstrate a novel technique for the microscopic patterning of gold by combining the photoreduction of Au^(Ⅲ)Br_(2)^(-)to Au^(Ⅰ)Br_(2)^(-)and the electrochemical reduction of Au^(Ⅰ)Br_(2)^(-)to elemental gold in a single step within solution. While mask-based methods have been the norm for electroplating, the adoption of direct laser writing for flexible, real-time patterning has not been widespread. Through irradiation using a 405 nm laser and applying a voltage corresponding to a selective potential window specific to Au^(Ⅰ)Br_(2)^(-), we have shown that we can locally deposit elemental gold at the focal point of the laser. In addition to demonstrating the feasibility of the technique, we have collected data on the kinetics of the photoreduction reaction in ethanol and have deduced its rate law. We have confirmed the selective deposition of Au^(Ⅰ)Br_(2)^(-) within a potential window through controlled potential electrolysis experiments and through direct measurement on a quartz crystal microbalance. Finally, we have verified local deposition through scanning electron microscopy.

Use of the GORE^(®) DrySeal Flex Introducer Sheath to Facilitate Implantation of the Transcatheter Venus P-valve

Objectives:We report our experience of using the 65 cm large diameter GORE^(®) DrySeal Flex Introducer sheath to facilitate transcatheter implantation of the Venus P-valve in the pulmonary position.Background:Transcatheter implantation of pulmonary valves can be difficult due to rigidity of the valve delivery system or the anatomy of the RVOT and pulmonary artery bifurcation and the risk of iatrogenic damage to the tricuspid valve support apparatus.Using long sheaths to pass and protect the tricuspid valve may facilitate the procedure.Methods:Multi-centre registry of patients who underwent transcatheter pulmonary valve implantation of the Venus P-valve using the GORE^(®) DrySeal Flex introducer sheath to facilitate passage of the valve to the right ventricular outflow tract.Procedural success,time to valve implantation and radiation safety parameters were analyzed.These data were compared to a control group of subjects treated between July 2014 and May 2016 with the same valve but without the use of GORE^(®) DrySeal.Results:Between December 2016 and September 2018,the Venus P-valve was successfully deployed through the GORE^(®) DrySeal in 12 patients.There were no procedure-related complications.As a control group,10 subjects treated between July 2014 and May 2016 were included.Total procedure time was significantly shorter in the GORE^(®) DrySeal group compared to the control group 96±27 min vs.164±12 min(p<0.001).Total screening time was significantly shorter in the GORE^(®) DrySeal group(24±11 min)when compared with the control group(32±14 min,p<0.001).Conclusions:We describe a modification to the previously described techniques of implanting the Venus P-valve in the pulmonary position after surgical repair of congenital heart disease.In our experience,the GORE^(®) DrySeal sheath has considerably facilitated the procedure and reduced the potential risks.