Enhanced Detection of Heart Valve Disease Using Integrated Artificial Intelligence at Scale.

BACKGROUND: Undertreatment of heart valve disease creates unnecessary patient risk. Poorly integrated healthcare data systems are unequipped to solve this problem. A software program using a rules-based algorithm to search the electronic health record for heart valve disease among patients treated by healthcare systems in the United States may provide a solution. METHODS: A software interface allowed concurrent access to picture archiving communication systems, the electronic health record, and other sources. The software platform was created to programmatically run a rules engine to search structured and unstructured data for identification of moderate or severe heart valve disease using guideline-reported values. Incidence and progression of disease as well as compliance with a care pathway were assessed. RESULTS: In 2 health institutions in the United States 60,145 patients had 77,215 echocardiograms. Moderate or severe aortic stenosis (AS) was identified at a rate of 9.1% of patients (5474 and 6910 echocardiograms) in this population. The precision and accuracy of the algorithm for the detection of moderate or severe AS was 92.9% and 98.6%, respectively. Thirty-five percent of patients (441/1265) with moderate stenosis and a subsequent echocardiogram progressed to severe stenosis (mean interval, 358 days). In 1 sample 70.3% of moderate AS patients lacked a 6-month echocardiogram or appointment. The platform enabled 100% accountability for all patients with severe AS. CONCLUSIONS: A rules-based software program enhances detection of heart valve disease and can be used to measures disease progression and care pathway compliance.

Changes in Cortical Activity in Stroke Survivors Undergoing Botulinum Neurotoxin Therapy for Treatment of Focal Spasticity.

Background: Botulinum NeuroToxin-A (BoNT-A) relieves muscle spasticity and increases range of motion necessary for stroke rehabilitation. Determining the effects of BoNT-A therapy on brain neuroplasticity could help physicians customize its use and predict its outcome. Objective: The purpose of this study was to investigate the effects of Botulinum Toxin-A therapy for treatment of focal spasticity on brain activation and functional connectivity. Design: We used functional Magnetic Resonance Imaging (fMRI) to track changes in blood oxygen-level dependent (BOLD) activation and functional connectivity associated with BoNT-A therapy in nine chronic stroke participants, and eight age-matched controls. Scans were acquired before BoNT-A injections (W0) and 6 weeks after the injections (W6). The task fMRI scan consisted of a block design of alternating mass finger flexion and extension. The voxel-level changes in BOLD activation, and pairwise changes in functional connectivity were analyzed for BoNT-A treatment (stroke W0 vs. W6). Results: BoNT-A injection therapy resulted in significant increases in brain activation in the contralesional premotor cortex, cingulate gyrus, thalamus, superior cerebellum, and in the ipsilesional sensory integration area. Lastly, cerebellar connectivity correlated with the Fugl-Meyer assessment of motor impairment before injection, while premotor connectivity correlated with the Fugl-Meyer score after injection. Conclusion: BoNT-A therapy for treatment of focal spasticity resulted in increased brain activation in areas associated with motor control, and cerebellar connectivity correlated with motor impairment before injection. These results suggest that neuroplastic effects might take place in response to improvements in focal spasticity.

Indirect Structural Connectivity Identifies Changes in Brain Networks After Stroke.

Background/Purpose: The purpose of this study was (1) to identify changes in structural connectivity after stroke and (2) to relate changes in indirect connectivity to post-stroke impairment. Methods: A novel measure of indirect connectivity was implemented to assess the impact of stroke on brain connectivity. Probabilistic tractography was performed on 13 chronic stroke and 16 control participants to estimate connectivity between gray matter (GM) regions. The Fugl-Meyer assessment of motor impairment was measured for stroke participants. Network measures of direct and indirect connectivity were calculated, and these measures were linearly combined with measures of white matter integrity to predict motor impairment. Results: We found significantly reduced indirect connectivity in the frontal and parietal lobes, ipsilesional subcortical regions, and bilateral cerebellum after stroke. When added to the regression analysis, the volume of GM with reduced indirect connectivity significantly improved the correlation between image parameters and upper extremity motor impairment (R2 = 0.71, p < 0.05). Conclusion: This study provides evidence of changes in indirect connectivity in regions remote from the lesion, particularly in the cerebellum and regions in the fronto-parietal cortices, and these changes correlate with upper extremity motor impairment. These results highlight the value of using measures of indirect connectivity to identify the effect of stroke on brain networks. Impact statement Changes in indirect structural connectivity occur in regions distant from a lesion after stroke, highlighting the impact that stroke has on brain functional networks. Specifically, losses in indirect structural connectivity occur in hubs with high centrality, including the fronto-parietal cortices and cerebellum. These losses in indirect connectivity more accurately reflect motor impairments than measures of direct structural connectivity. As a consequence, indirect structural connectivity appears to be important to recovery after stroke and imaging biomarkers that incorporate indirect structural connectivity might improve prognostication of stroke outcomes.

Tasked-Based Functional Brain Connectivity in Multisensory Control of Wrist Movement After Stroke.

In this study we documented brain connectivity associated with multisensory integration during wrist control in healthy young adults, aged matched controls and stroke survivors. A novel functional MRI task paradigm involving wrist movement was developed to gain insight into the effects of multimodal sensory feedback on brain functional networks in stroke participants. This paradigm consisted of an intermittent position search task using the wrist during fMRI signal acquisition with visual and auditory feedback of proximity to a target position. We enrolled 12 young adults, 10 participants with chronic post-stroke hemiparesis, and nine age-matched controls. Activation maps were obtained, and functional connectivity networks were calculated using an independent component analysis (ICA) approach. Task-based networks were identified using activation maps, and nodes were obtained from the ICA components. These nodes were subsequently used for connectivity analyses. Stroke participants demonstrated significantly greater contralesional activation than controls during the visual feedback condition and less ipsilesional activity than controls during the auditory feedback condition. The sensorimotor component obtained from the ICA differed between rest and task for control and stroke participants: task-related lateralization to the contralateral cortex was observed in controls, but not in stroke participants. Connectivity analyses between the lesioned sensorimotor cortex and the contralesional cerebellum demonstrated decreased functional connectivity in stroke participants (p < 0.005), which was positively correlated the Box and Blocks arm function test (r 2 = 0.59). These results suggest that task-based functional connectivity provides detail on changes in brain networks in stroke survivors. The data also highlight the importance of cerebellar connections for recovery of arm function after stroke.