Explainable artificial intelligence model to predict acute critical illness from electronic health records.

Acute critical illness is often preceded by deterioration of routinely measured clinical parameters, e.g., blood pressure and heart rate. Early clinical prediction is typically based on manually calculated screening metrics that simply weigh these parameters, such as early warning scores (EWS). The predictive performance of EWSs yields a tradeoff between sensitivity and specificity that can lead to negative outcomes for the patient. Previous work on electronic health records (EHR) trained artificial intelligence (AI) systems offers promising results with high levels of predictive performance in relation to the early, real-time prediction of acute critical illness. However, without insight into the complex decisions by such system, clinical translation is hindered. Here, we present an explainable AI early warning score (xAI-EWS) system for early detection of acute critical illness. xAI-EWS potentiates clinical translation by accompanying a prediction with information on the EHR data explaining it.

Listening to the patients: using participatory design in the development of a cardiac telerehabilitation web portal.

BACKGROUND: Cardiovascular disease is the leading cause of all deaths worldwide. Cardiac rehabilitation is an effective approach for preventing secondary complications, but it remains a complex intervention because of the need for lifestyle changes. One solution is to employ interactive telerehabilitation or eHealth web portals. However, these have not been implemented as intended by developers. The aim of this study was to evaluate the design and usability of a cardiac telerehabilitation web portal, called the 'HeartPortal', for use among heart failure (HF) patients. METHODS: The HeartPortal was designed using participatory design (PD). The design process involved HF patients, their relatives, healthcare professionals (HCP), healthcare company specialists and researchers. Self-determination theory (SDT) was used to enable the design to elicit intrinsic motivation within the patients. With eHealth literacy skills in mind, the goal of the HeartPortal was to successfully target the end-users. The PD process and data collection techniques included cultural probes, workshops, participant-observation, questionnaires, and problem-solving tasks. RESULTS: The PD process helped us design an interactive web portal, the HeartPortal. Based on participants' feedback, the design incorporated features such as being able to make notes and to communicate with HCP, view data from self-tracking devices in a graphic form, and to obtain information on rehabilitation in the form of text, audio, and video. More than half of those testing the HeartPortal found that it was easy to navigate, and most of the users stated that it had an excellent structure and that using it could possibly improve their condition. CONCLUSIONS: Overall, the HeartPortal was found to be logical and easy to navigate and will now be tested in a clinical trial within the Future Patient Telerehabilitation Program.

Altered excitability of small cutaneous nerve fibers during cooling assessed with the perception threshold tracking technique.

BACKGROUND: There is a need for new approaches to increase the knowledge of the membrane excitability of small nerve fibers both in healthy subjects, as well as during pathological conditions. Our research group has previously developed the perception threshold tracking technique to indirectly assess the membrane properties of peripheral small nerve fibers. In the current study, a new approach for studying membrane excitability by cooling small fibers, simultaneously with applying a slowly increasing electrical stimulation current, is evaluated. The first objective was to examine whether altered excitability during cooling could be detected by the perception threshold tracking technique. The second objective was to computationally model the underlying ionic current that could be responsible for cold induced alteration of small fiber excitability. The third objective was to evaluate whether computational modelling of cooling and electrical simulation can be used to generate hypotheses of ionic current changes in small fiber neuropathy. RESULTS: The excitability of the small fibers was assessed by the perception threshold tracking technique for the two temperature conditions, 20 °C and 32 °C. A detailed multi-compartment model was developed, including the ionic currents: NaTTXs, NaTTXr, NaP, KDr, KM, KLeak, KA, and Na/K-ATPase. The perception thresholds for the two long duration pulses (50 and 100 ms) were reduced when the skin temperature was lowered from 32 to 20 °C (p < 0.001). However, no significant effects were observed for the shorter durations (1 ms, p = 0.116; 5 ms p = 0.079, rmANOVA, Sidak). The computational model predicted that the reduction in the perception thresholds related to long duration pulses may originate from a reduction of the KLeak channel and the Na/K-ATPase. For short durations, the effect cancels out due to a reduction of the transient TTX resistant sodium current (Nav1.8). Additionally, the result from the computational model indicated that cooling simultaneously with electrical stimulation, may increase the knowledge regarding pathological alterations of ionic currents. CONCLUSION: Cooling may alter the ionic current during electrical stimulation and thereby provide additional information regarding membrane excitability of small fibers in healthy subjects and potentially also during pathological conditions.

Automatic Bone Marrow Cellularity Estimation in H&E Stained Whole Slide Images.

Bone marrow cellularity is an important measure in diagnostic hematopathology. Currently, the gold standard for bone marrow cellularity estimation is manual inspection of hematoxylin and eosin stained whole slide images (H&E WSI) by hematopathologists. However, these assessments are subjective and subject to interobserver and intraobserver variability. This may be reduced by using a computer-assisted estimate of bone marrow cellularity. The aim of this study was to develop a fully automated algorithm to estimate bone marrow cellularity in H&E WSI stains using bone marrow segmentation. Data consisted of eight bone marrow H&E WSIs extracted from eight subjects. An algorithm was developed to estimate the bone marrow cellularity consisting of biopsy segmentation, tissue classification, and bone marrow segmentation. Segmentations of the red and yellow bone marrow (YBM) were used to estimate the bone marrow cellularity within the WSI H&E stains. The DICE coefficient between automatic tissue segmentations and ground truth segmentations conducted by an experienced hematopathologist were used for validation. Furthermore, the agreement between the automatic and two manual cellularity estimates was assessed using Bland-Altman plots and intraclass correlation coefficients (ICC). The validation of the bone marrow segmentation demonstrated an average DICE of 0.901 and 0.920 for the red and YBM, respectively. A mean cellularity estimate difference of -0.552 and - 7.816 was obtained between the automatic cellularity estimates and two manual cellularity estimates, respectively. An ICC of 0.980 (95%CI: 0.925-0.995, P-value: 5.51 × 10-7 ) was obtained between the automatic and manual cellularity estimates based on manual annotations. The study demonstrated that it was possible to obtain bone marrow cellularity estimates with a good agreement with bone marrow cellularity estimates obtained from an experienced hematopathologist. © 2019 International Society for Advancement of Cytometry.

SHARQnet - Sophisticated harmonic artifact reduction in quantitative susceptibility mapping using a deep convolutional neural network.

Quantitative susceptibility mapping (QSM) reveals pathological changes in widespread diseases such as Parkinson's disease, Multiple Sclerosis, or hepatic iron overload. QSM requires multiple processing steps after the acquisition of magnetic resonance imaging (MRI) phase measurements such as unwrapping, background field removal and the solution of an ill-posed field-to-source-inversion. Current techniques utilize iterative optimization procedures to solve the inversion and background field correction, which are computationally expensive and lead to suboptimal or over-regularized solutions requiring a careful choice of parameters that make a clinical application of QSM challenging. We have previously demonstrated that a deep convolutional neural network can invert the magnetic dipole kernel with a very efficient feed forward multiplication not requiring iterative optimization or the choice of regularization parameters. In this work, we extended this approach to remove background fields in QSM. The prototype method, called SHARQnet, was trained on simulated background fields and tested on 3T and 7T brain datasets. We show that SHARQnet outperforms current background field removal procedures and generalizes to a wide range of input data without requiring any parameter adjustments. In summary, we demonstrate that the solution of ill-posed problems in QSM can be achieved by learning the underlying physics causing the artifacts and removing them in an efficient and reliable manner and thereby will help to bring QSM towards clinical applications.