Abstract-We compare the performance of five indices of retinal vessel tortuosity against sampling rates of vessel centerlines. We consider distance measure, tortuosity density, two curvature-based measures, and a recently introduced slopechain coding for general curves, never before assessed comparatively with retinal vessels. To enable replication of our results, we use the public dataset for retinal tortuosity, RET-TORT. We find that (1) the tortuosity density index offers good performance overall, but is not always the best performer; (2) curvature-based methods are the best if high-frequency resampling is possible, but (3) are the most sensitive to variations of the number of samples; (4) slope-chain coding performs best at low sampling rates, but the length of the linear elements must be chosen carefully. In general, performance may vary considerably with resampling, suggesting that the choice of a tortuosity index for clinical inference requires attention to numerical details, and ideally standardization thereof.

Objective: This work investigates the use of deep convolutional neural networks (CNN) to automatically perform measurements of fetal body parts, including head circumference, biparietal diameter, abdominal circumference and femur length, and to estimate gestational age and fetal weight using fetal ultrasound videos. Approach: We developed a novel multi-task CNN-based spatio-temporal fetal US feature extraction and standard plane detection algorithm (called FUVAI) and evaluated the method on 50 freehand fetal US video scans. We compared FUVAI fetal biometric measurements with measurements made by five experienced sonographers at two time points separated by at least two weeks. Intra- and inter-observer variabilities were estimated. Main Results: We found that automated fetal biometric measurements obtained by FUVAI were comparable to the measurements performed by experienced sonographers The observed differences in measurement values were within the range of inter- and intra-observer variability. Moreover, analysis has shown that these differences were not statistically significant when comparing any individual medical expert to our model. Significance: We argue that FUVAI has the potential to assist sonographers who perform fetal biometric measurements in clinical settings by providing them with suggestions regarding the best measuring frames, along with automated measurements. Moreover, FUVAI is able perform these tasks in just a few seconds, which is a huge difference compared to the average of six minutes taken by sonographers. This is significant, given the shortage of medical experts capable of interpreting fetal ultrasound images in numerous countries.

We present a semantically interpretable system for automated ICD coding of clinical text documents. Our contribution is an ontological attention mechanism which matches the structure of the ICD ontology, in which shared attention vectors are learned at each level of the hierarchy, and combined into label-dependent ensembles. Analysis of the attention heads shows that shared concepts are learned by the lowest common denominator node. This allows child nodes to focus on the differentiating concepts, leading to efficient learning and memory usage. Visualisation of the multilevel attention on the original text allows explanation of the code predictions according to the semantics of the ICD ontology. On the MIMIC-III dataset we achieve a 2.7% absolute (11% relative) improvement from 0.218 to 0.245 macro-F1 score compared to the previous state of the art across 3,912 codes. Finally, we analyse the labelling inconsistencies arising from different coding practices which limit performance on this task.

No abstract