Automated polyp detection in the colorectum: a prospective study (with videos).

BACKGROUND AND AIMS: Adenoma detection is a highly personalized task that differs markedly among endoscopists. Technical advances are therefore desirable for the improvement of the adenoma detection rate (ADR). An automated computer-driven technology would offer the chance to objectively assess the presence of colorectal polyps during colonoscopy. We present here the application of a real-time automated polyp detection software (APDS) under routine colonoscopy conditions. METHODS: This was a prospective study at a university hospital in Germany. A prototype of a novel APDS ("KoloPol," Fraunhofer IIS, Erlangen, Germany) was used for automated image-based polyp detection. The software functions by highlighting structures of possible polyp lesions in a color-coded manner during real-time colonoscopy procedures. Testing the feasibility of APDS deployment under real-time conditions was the primary goal of the study. APDS polyp detection rates (PDRs) were defined as secondary endpoints provided that endoscopists' detection served as criterion standard. RESULTS: The APDS was applied in 55 routine colonoscopies without the occurrence of any clinically relevant adverse events. Endoscopists' PDRs and ADRs were 56.4% and 30.9%, respectively. The PDRs and ADRs of the APDS were 50.9% and 29.1%, respectively. The APDS detected 55 of 73 polyps (75.3%). Smaller polyp size and flat polyp morphology were correlated with insufficient polyp detection by the APDS. CONCLUSION: Computer-assisted automated low-delay polyp detection is feasible during real-time colonoscopy. Efforts should be undertaken to improve the APDS with respect to smaller and flat shaped polyps. (Clinical trial registration number: NCT02838888.).

Using simulated fluorescence cell micrographs for the evaluation of cell image segmentation algorithms.

BACKGROUND: Manual assessment and evaluation of fluorescent micrograph cell experiments is time-consuming and tedious. Automated segmentation pipelines can ensure efficient and reproducible evaluation and analysis with constant high quality for all images of an experiment. Such cell segmentation approaches are usually validated and rated in comparison to manually annotated micrographs. Nevertheless, manual annotations are prone to errors and display inter- and intra-observer variability which influence the validation results of automated cell segmentation pipelines. RESULTS: We present a new approach to simulate fluorescent cell micrographs that provides an objective ground truth for the validation of cell segmentation methods. The cell simulation was evaluated twofold: (1) An expert observer study shows that the proposed approach generates realistic fluorescent cell micrograph simulations. (2) An automated segmentation pipeline on the simulated fluorescent cell micrographs reproduces segmentation performances of that pipeline on real fluorescent cell micrographs. CONCLUSION: The proposed simulation approach produces realistic fluorescent cell micrographs with corresponding ground truth. The simulated data is suited to evaluate image segmentation pipelines more efficiently and reproducibly than it is possible on manually annotated real micrographs.