Deep Learning Approach for Automatic Microaneurysms Detection

In diabetic retinopathy (DR), the early signs that may lead the eyesight towards complete vision loss are considered as microaneurysms (MAs). The shape of these MAs is almost circular, and they have a darkish color and are tiny in size, which means they may be missed by manual analysis of ophthalmologists. In this case, accurate early detection of microaneurysms is helpful to cure DR before non-reversible blindness. In the proposed method, early detection of MAs is performed using a hybrid feature embedding approach of pre-trained CNN models, named as VGG-19 and Inception-v3. The performance of the proposed approach was evaluated using publicly available datasets, namely “E-Ophtha” and “DIARETDB1”, and achieved 96% and 94% classification accuracy, respectively. Furthermore, the developed approach outperformed the state-of-the-art approaches in terms of sensitivity and specificity for microaneurysms detection.

dsSurvival: Privacy preserving survival models for federated individual patient meta-analysis in DataSHIELD

Objective Achieving sufficient statistical power in a survival analysis usually requires large amounts of data from different sites. Sensitivity of individual-level data, ethical and practical considerations regarding data sharing across institutions could be a potential challenge for achieving this added power. Hence we implemented a federated meta-analysis approach of survival models in DataSHIELD, where only anonymous aggregated data are shared across institutions, while simultaneously allowing for exploratory, interactive modelling. In this case, meta-analysis techniques to combine analysis results from each site are a solution, but a manual analysis workflow hinders exploration. Thus, the aim is to provide a framework for performing meta-analysis of Cox regression models across institutions without manual analysis steps for the data providers. Results We introduce a package ( dsSurvival) which allows privacy preserving meta-analysis of survival models, including the calculation of hazard ratios. Our tool can be of great use in biomedical research where there is a need for building survival models and there are privacy concerns about sharing data.

dsSurvival: Privacy preserving survival models for federated individual patient meta-analysis in DataSHIELD

Achieving sufficient statistical power in a survival analysis usually requires large amounts of data from different sites. Sensitivity of individual-level data, ethical and practical considerations regarding data sharing across institutions could be a potential challenge for achieving this added power. Hence we implemented a federated meta-analysis approach of survival models in DataSHIELD, where only anonymous aggregated data are shared across institutions, while simultaneously allowing for exploratory, interactive modelling. In this case, meta-analysis techniques to combine analysis results from each site are a solution, but a manual analysis workflow hinders exploration. Thus, the aim is to provide a framework for performing meta-analysis of Cox regression models across institutions without manual analysis steps for the data providers. We introduce a package (dsSurvival) which allows privacy preserving meta-analysis of survival models, including the calculation of hazard ratios. Our tool can be of great use in biomedical research where there is a need for building survival models and there are privacy concerns about sharing data. A tutorial in bookdown format with code, diagnostics, plots and synthetic data is available here: https://neelsoumya.github.io/dsSurvivalbookdown/ All code is available from the following repositories: https://github.com/neelsoumya/dsSurvivalClient/ https://github.com/neelsoumya/dsSurvival/

Monitoring Services to Support Continuous Curriculum Engineering

Continuous and rapid developments in science and technology have raised a challenge for compliance between the learning content provided by study programs and (1) the actual state of art in the domains the programs are addressing and (2) the actual needs of companies that will employ the graduates of these programs. To achieve such compliance continuously, the digitalization of learning content (curriculum) engineering could provide supporting tools that facilitate awareness of incompliance, which is the first step in introducing changes in the learning content. In this article we discuss how such awareness, regarding the needs of companies, can be supported by a service system that monitors the gap between educational demand and offer. The proposed service system provides possibilities for automatic, semi-automatic and manual analysis of texts representing the educational demand, versus the texts representing the educational offer. The implementation of the service system has been demonstrated, having been applied at university. The experiments showed that the system can provide valuable information for educational content development, but its maintenance and incorporation in study process management and curriculum engineering still require additional research.

DeepACSA: Automatic segmentation of anatomical cross-sectional area in ultrasound images of human lower limb muscles using deep learning

Background: Muscle anatomical cross-sectional area (ACSA) is an important parameter that characterizes muscle function and helps to classify the severity of several muscular disorders. Ultrasound is a patient friendly, fast and cheap method of assessing muscle ACSA, but manual analysis of the images is laborious, subjective and requires thorough experience. To date, no open access and fully automated program to segment ACSA in ultrasound images is available. On this basis, we present DeepACSA, a deep learning approach to automatically segment ACSA in panoramic ultrasound images of the human rectus femoris (RF), vastus lateralis (VL), gastrocnemius medialis (GM) and lateralis (GL) muscles. Methods: We trained convolutional neural networks using 1772 ultrasound images from 153 participants (25 females, 128 males; mean age = 38.2 years, range: 13-78) captured by three experienced operators using three distinct devices. We trained three muscle-specific models to detect ACSA. Findings: Comparing DeepACSA analysis of the RF to manual analysis resulted in intra-class correlation (ICC) of 0.96 (95% CI 0.94,0.97), mean difference of 0.31 cm2 (0.04,0.58) and standard error of the differences (SEM) of 0.91 cm2 (0.47,1.36). For the VL, ICC was 0.94 (0.91,0.96), mean difference was 0.25 cm2 (-0.21,0.7) and SEM was 1.55 cm2 (1.13,1.96). The GM/GL muscles demonstrated an ICC of 0.97 (0.95,0.98), a mean difference of 0.01 cm2 (-0.25, 0.24) and a SEM of 0.69 cm2 (0.52,0.83). Interpretation: DeepACSA provides fast and objective segmentation of lower limb panoramic ultrasound images comparable to manual segmentation and is easy to implement both in research and clinical settings. Inaccurate model predictions occurred predominantly on low-quality images, highlighting the importance of high image quality for accurate prediction.

MR elastography in nonalcoholic fatty liver disease: inter-center and inter-analysis-method measurement reproducibility and accuracy at 3T

Objectives To assess reproducibility and fibrosis classification accuracy of magnetic resonance elastography (MRE)–determined liver stiffness measured manually at two different centers, and by automated analysis software in adults with nonalcoholic fatty liver disease (NAFLD), using histopathology as a reference standard. Methods This retrospective, cross-sectional study included 91 adults with NAFLD who underwent liver MRE and biopsy. MRE-determined liver stiffness was measured independently for this analysis by an image analyst at each of two centers using standardized manual analysis methodology, and separately by an automated analysis. Reproducibility was assessed pairwise by intraclass correlation coefficient (ICC) and Bland-Altman analysis. Diagnostic accuracy was assessed by receiver operating characteristic (ROC) analyses. Results ICC of liver stiffness measurements was 0.95 (95% CI: 0.93, 0.97) between center 1 and center 2 analysts, 0.96 (95% CI: 0.94, 0.97) between the center 1 analyst and automated analysis, and 0.94 (95% CI: 0.91, 0.96) between the center 2 analyst and automated analysis. Mean bias and 95% limits of agreement were 0.06 ± 0.38 kPa between center 1 and center 2 analysts, 0.05 ± 0.32 kPa between the center 1 analyst and automated analysis, and 0.11 ± 0.41 kPa between the center 2 analyst and automated analysis. The area under the ROC curves for the center 1 analyst, center 2 analyst, and automated analysis were 0.834, 0.833, and 0.847 for distinguishing fibrosis stage 0 vs. ≥ 1, and 0.939, 0.947, and 0.940 for distinguishing fibrosis stage ≤ 2 vs. ≥ 3. Conclusion MRE-determined liver stiffness can be measured with high reproducibility and fibrosis classification accuracy at different centers and by an automated analysis. Key Points • Reproducibility of MRE liver stiffness measurements in adults with nonalcoholic fatty liver disease is high between two experienced centers and between manual and automated analysis methods. • Analysts at two centers had similar high diagnostic accuracy for distinguishing dichotomized fibrosis stages. • Automated analysis provides similar diagnostic accuracy as manual analysis for advanced fibrosis.

Mobile Eye-Tracking Data Analysis Using Object Detection via YOLO v4

Remote eye tracking has become an important tool for the online analysis of learning processes. Mobile eye trackers can even extend the range of opportunities (in comparison to stationary eye trackers) to real settings, such as classrooms or experimental lab courses. However, the complex and sometimes manual analysis of mobile eye-tracking data often hinders the realization of extensive studies, as this is a very time-consuming process and usually not feasible for real-world situations in which participants move or manipulate objects. In this work, we explore the opportunities to use object recognition models to assign mobile eye-tracking data for real objects during an authentic students’ lab course. In a comparison of three different Convolutional Neural Networks (CNN), a Faster Region-Based-CNN, you only look once (YOLO) v3, and YOLO v4, we found that YOLO v4, together with an optical flow estimation, provides the fastest results with the highest accuracy for object detection in this setting. The automatic assignment of the gaze data to real objects simplifies the time-consuming analysis of mobile eye-tracking data and offers an opportunity for real-time system responses to the user’s gaze. Additionally, we identify and discuss several problems in using object detection for mobile eye-tracking data that need to be considered.

Healthcare professionals’ perceptions on medication communication challenges and solutions – text mining and manual content analysis - cross-sectional study

Background Communication challenges contribute to medication incidents in hospitals, but it is unclear how communication can be improved. The aims of this study were threefold: firstly, to describe the most common communication challenges related to medication incidents as perceived by healthcare professionals across specialized hospitals for adult patients; secondly, to consider suggestions from healthcare professionals with regard to improving medication communication; and thirdly, to explore how text mining compares to manual analysis when analyzing the free-text content of survey data. Methods This was a cross-sectional, descriptive study. A digital survey was sent to professionals in two university hospital districts in Finland from November 1, 2019, to January 31, 2020. In total, 223 professionals answered the open-ended questions; respondents were primarily registered nurses (77.7 %), physicians (8.6 %), and pharmacists (7.3 %). Text mining and manual inductive content analysis were employed for qualitative data analysis. Results The communication challenges were: (1) inconsistent documentation of prescribed and administered medication; (2) failure to document orally given prescriptions; (3) nurses’ unawareness of prescriptions (given outside of ward rounds) due to a lack of oral communication from the prescribers; (4) breaks in communication during care transitions to non-communicable software; (5) incomplete home medication reconciliation at admission and discharge; (6) medication lists not being updated during the inpatient period due to a lack of clarity regarding the responsible professional; and (7) work/environmental factors during medication dispensation and the receipt of verbal prescriptions. Suggestions for communication enhancements included: (1) structured digital prescriptions; (2) guidelines and training on how to use documentation systems; (3) timely documentation of verbal prescriptions and digital documentation of administered medication; (4) communicable software within and between organizations; (5) standardized responsibilities for updating inpatients’ medication lists; (6) nomination of a responsible person for home medication reconciliation at admission and discharge; and (7) distraction-free work environment for medication communication. Text mining and manual analysis extracted similar primary results. Conclusions Non-communicable software, non-standardized medication communication processes, lack of training on standardized documentation, and unclear responsibilities compromise medication safety in hospitals. Clarification is needed regarding interdisciplinary medication communication processes, techniques, and responsibilities. Text mining shows promise for free-text analysis.

Giant Unilamellar Vesicle Electroformation: What to Use, What to Avoid, and How to Quantify the Results

Since its inception more than thirty years ago, electroformation has become the most commonly used method for growing giant unilamellar vesicles (GUVs). Although the method seems quite straightforward at first, researchers must consider the interplay of a large number of parameters, different lipid compositions, and internal solutions in order to avoid artifactual results or reproducibility problems. These issues motivated us to write a short review of the most recent methodological developments and possible pitfalls. Additionally, since traditional manual analysis can lead to biased results, we have included a discussion on methods for automatic analysis of GUVs. Finally, we discuss possible improvements in the preparation of GUVs containing high cholesterol contents in order to avoid the formation of artifactual cholesterol crystals. We intend this review to be a reference for those trying to decide what parameters to use as well as an overview providing insight into problems not yet addressed or solved.

Updated GNSS velocity solution in the Nordic and Baltic countries with a semi-automatic offset detection method

In Fennoscandia, the Glacial Isostatic Adjustment (GIA) causes intraplate deformations that affect the national static reference frames. The GNSS-determined velocities are important data for constraining the GIA models, which are necessary for maintaining the national reference frames. The Nordic Geodetic Commission (NKG) has published a dense and consistent GNSS station velocity solution in 2019, and we present now an update of the solution covering additional 3.5 years of data. Undetected positional offsets are the main factor decreasing the accuracy of the velocity estimates. We developed a method for the semi-automatic offset detection to improve the quality of our solution. The results show that we could correctly detect 74% of the manually determined offsets, and the undetected offsets would have caused a median 0.1 mm/y bias in trend. The method pointed out some otherwise unnoticed offsets and will decrease the need for manual analysis in the future. The updated velocity solution especially improves the velocity estimates of the newly established stations and the quality of the velocity estimates in Baltic countries. The formal uncertainties estimated using the power-law plus white noise model were at a median of 0.06 and 0.15 mm/y for horizontal and vertical velocities, respectively. However, we concluded that the systematic velocity uncertainties due to the reference frame alignment were approximately at the same level.