Machine Learning for Predicting Motor Improvement After Acute Subcortical Infarction Using Baseline Whole Brain Volumes

Background. Neuroimaging biomarkers are valuable predictors of motor improvement after stroke, but there is a gap between published evidence and clinical usage. Objective. In this work, we aimed to investigate whether machine learning techniques, when applied to a combination of baseline whole brain volumes and clinical data, can accurately predict individual motor outcome after stroke. Methods. Upper extremity Fugl-Meyer Assessments (FMA-UE) were conducted 1 week and 12 weeks, and structural MRI was performed 1 week, after onset in 56 patients with subcortical infarction. Proportional recovery model residuals were employed to assign patients to proportional and poor recovery groups (34 vs 22). A sophisticated machine learning scheme, consisting of conditional infomax feature extraction, synthetic minority over-sampling technique for nominal and continuous, and bagging classification, was employed to predict motor outcomes, with the input features being a combination of baseline whole brain volumes and clinical data (FMA-UE scores). Results. The proposed machine learning scheme yielded an overall balanced accuracy of 87.71% in predicting proportional vs poor recovery outcomes, a sensitivity of 93.77% in correctly identifying poor recovery outcomes, and a ROC AUC of 89.74%. Compared with only using clinical data, adding whole brain volumes can significantly improve the classification performance, especially in terms of the overall balanced accuracy (from 80.88% to 87.71%) and the sensitivity (from 92.23% to 93.77%). Conclusions. Experimental results suggest that a combination of baseline whole brain volumes and clinical data, when equipped with appropriate machine learning techniques, may provide valuable information for personalized rehabilitation planning after subcortical infarction.

Association of neutrophil-to-lymphocyte ratio (NLR) with the prognosis of first attack neuromyelitis optica spectrum disorder (NMOSD): a retrospective cohort study

Background To investigate the relationship between the neutrophil-to-lymphocyte ratio (NLR) and prognosis after the first attack of optic neuromyelitis optica spectrum disorder (NMOSD). Methods In this retrospective study, we included the medical records of 324 patients with first episode NMOSD and collected data on clinical parameters. Follow-up extended disability status scale (EDSS) score and relapse rate were analyzed using logistic regression models to determine the independent effect of NLR on outcomes; receiver operating characteristic (ROC) curves were applied to analyze the predictive value of NLR for the prognosis of NMOSD. Interaction and stratification analyses were used to explore the association between NLR and prognosis of patients with NMOSD, and Kaplan-Meier analysis was used to investigate the relationship between NLR and outcome. The association between NLR level with relapse rate and poor recovery was assessed by a Cox regression analysis. Results Patients in the high-NLR group had significantly higher EDSS scores and relapse rates at follow-up (both, P < 0.001) than did those in the low-NLR group. Univariate analysis showed revealed that NLR was significantly associated with relapse (odds ratio [OR] = 1.28, 95% confidence interval [CI]: 1.16–1.41, P < 0.001) and poor recovery (OR = 1.32, 95% CI: 1.20–1.46, P < 0.001), and these associations remained significant, even after multifactorial analysis (OR = 1.33, 95% CI: 1.11–1.59, P = 0.002; OR = 1.23, 95% CI: 1.06–1.43, P = 0.007, respectively). Stratified analysis showed that sex, platelet-to-lymphocyte ratio (PLR) level, and lymphocyte-to-monocyte technical ratio (LMR) level were strongly associated with relapse owing to elevated NLR; Kaplan-Meier survival curve analysis showed that the median time to relapse was significantly lower in the high-NLR group than in the low-NLR group (P < 0.001). A multivariate analysis showed a significant relationship between NLR level with relapse (HR = 1.07, 95%CI: 1.03–1.10, P = 0.001) and poor recovery (HR = 1.08, 95%CI: 1.04–1.11, P = 0.001). Conclusions NLR may be used as a prognostic indicator for first onset NMOSD, and a high NLR may be significantly associated with high relapse rates and poor recovery.

Lack of Glutamate Receptor Subunit Expression Changes in Hippocampal Dentate Gyrus after Experimental Traumatic Brain Injury in a Rodent Model of Depression

Traumatic brain injury (TBI) affects over 69 million people annually worldwide, and those with pre-existing depression have worse recovery. The molecular mechanisms that may contribute to poor recovery after TBI with co-morbid depression have not been established. TBI and depression have many commonalities including volume changes, myelin disruption, changes in proliferation, and changes in glutamatergic signaling. We used a well-established animal model of depression, the Wistar Kyoto (WKY) rat, to elucidate changes after TBI that may influence the recovery trajectory. We compared the histological and molecular outcomes in the hippocampal dentate gyrus after experimental TBI using the lateral fluid percussion injury (LFPI) in the WKY and the parent Wistar (WIS) strain. We showed that WKY had exaggerated myelin loss after LFPI and baseline deficits in proliferation. In addition, we showed that while after LFPI WIS rats exhibited glutamate receptor subunit changes, namely increased GluN2B, the WKY rats failed to show such injury-related changes. These differential responses to LFPI helped to elucidate the molecular characteristics that influence poor recovery after TBI in those with pre-existing depression and may lead to targets for future therapeutic interventions.

Postoperative Recovery Quality after General Anesthesia in Patients Undergoing Emergency Surgery

BACKGROUND: The quality of post-operative patient recovery is part of the indicators of the quality of anesthesia services. However, side effects were still found in post-operative patients, which resulted in the speed and quality of recovery (QoR) and an increase in the length of stay for patients in the hospital. AIM: This study aims to describe the quality of a patient’s recovery after general anesthesia for patients undergoing surgery in the emergency room. MATERIALS AND METHODS: This research is a descriptive study with a cross-sectional design conducted from February to July 2020. Patients numbered 52 people who are patients aged 18–59 years who are willing to fill out a questionnaire QoR-40 on the day after surgery. All data were analyzed using the statistical program SPSS. The median of the QoR-40 score will be used as a cutoff point between good and poor recovery quality. RESULTS: Fifty-two patients responded to the study with a median age of 44 years. The median QoR-40 score was 188, with most of the good category’s recovery quality (55.8%). The middle-aged group showed better QoR (68.2 vs. 31.8%), as well as the female compared to male (65 vs. 35%). Patients who underwent neurosurgery showed poor recovery 61.9 vs. 38.1%). CONCLUSION: More than half of the patients who underwent emergency surgery under general anesthesia in the emergency room of Sanglah Hospital have good recovery quality.

Antineutrophil Cytoplasmic Antibody-Associated Vasculitis With Acute Kidney Injury: Short-Term Recovery Predicts Long-Term Outcome

BackgroundKidney involvement is common in antineutrophil cytoplasmic antibody (ANCA) associated vasculitis (AAV). It tends to be aggressive, and in some patients, the kidney involvement may reach the criteria of acute kidney injury (AKI). Here, we aim to describe the clinical characteristics of these patients and find risk factors for poor outcomes.MethodsPatients diagnosed with AAV in our hospital from February 2003 to February 2017 were included. Those who reached the KDIGO AKI criteria were reclassified according to the KDIGO AKI stage. The clinical features of these patients were analyzed. Also, according to the variation of serum creatinine 3 months after AKI episode, patients were further divided into two groups: patients whose serum creatinine (Scr) level at the third month decreased by 30% or more from the peak Scr level was classified into G1 and others were classified into G2. Long-term renal and survival outcomes of these patients were analyzed with a Cox model. The renal endpoint was reaching end-stage renal disease (ESRD), and the survival endpoint was death. Nomograms were built based on cox models.ResultsOf 141 AAV patients included, during the median follow-up period of 64.0 (IQR 34.8, 85.4) months, 36 (25.5%) patients reached renal endpoints, and 22 (15.6%) patients died. The median renal survival time was 35.9 (IQR 21.3, 72.6) months and the median survival time was 48.4 (IQR 26.8, 82.8) months. Multivariate analysis showed that poor recovery of Scr level at 90 days (P &lt; 0.001, RR = 9.150, 95%CI 4.163–20.113), BVAS score (P = 0.014, RR = 1.110, 95% CI1.021–1.207), and AKI stage 3 (P = 0.012 RR = 3.116, 95%CI 1.278–7.598) were independent risk factors for renal endpoints; poor recovery of Scr level at 90 days (P = 0.010, RR = 3.264, 95%CI 1.326–8.035), BVAS score (P = 0.010, RR = 1.171, 95%CI 1.038–1.320) and age (P = 0.017, RR = 1.046, 95%CI 1.008–1.086) were independent risk factors for all-cause death. The c-index of nomograms is 0.830 for the renal outcome and 0.763 for the survival outcome.ConclusionKDIGO AKI stage 3 is the risk factor for ESRD in AAV patients with AKI. The BVAS score and level of kidney function recovery at 90 days are the independent risk factors for both ESRD and all-cause death and are of predictive value for the outcome.

Limited predictive value of illness perceptions for short-term poor recovery in musculoskeletal pain. A multi-center longitudinal study

Background Musculoskeletal pain (MSP) is recognized worldwide as a major cause of increased years lived with disability. In addition to known generic prognostic factors, illness perceptions (IPs) may have predictive value for poor recovery in MSP. We were interested in the added predictive value of baseline IPs, over and above the known generic prognostic factors, on clinical recovery from MSP. Also, it is hypothesized there may be overlap between IPs and domains covered by the Four-Dimensional Symptom Questionnaire (4DSQ), measuring distress, depression, anxiety and somatization. The aim of this study is twofold; 1) to assess the added predictive value of IPs for poor recovery and 2) to assess differences in predictive value for poor recovery between the Brief Illness Perception Questionnaire - Dutch Language Version (Brief IPQ-DLV) and the 4DSQ. Methods An eligible sample of 251 patients with musculoskeletal pain attending outpatient physical therapy were included in a multi-center longitudinal cohort study. Pain intensity, physical functioning and Global Perceived Effect were the primary outcomes. Hierarchical logistic regression models were used to assess the added value of baseline IPs for predicting poor recovery. To investigate the performance of the models, the levels of calibration (Hosmer-Lemeshov test) and discrimination (Area under the Curve (AUC)) were assessed. Results Baseline ‘Treatment Control’ added little predictive value for poor recovery in pain intensity [Odds Ratio (OR) 0.80 (Confidence Interval (CI) 0.66–0.97), increase in AUC 2%] and global perceived effect [OR 0.78 (CI 0.65–0.93), increase in AUC 3%]. Baseline ‘Timeline’ added little predictive value for poor recovery in physical functioning [OR 1.16 (CI 1.03–1.30), increase in AUC 2%]. There was a non-significant difference between AUCs in predictive value for poor recovery between the Brief IPQ-DLV and the 4DSQ. Conclusions Based on the findings of this explorative study, assessing baseline IPs, over and above the known generic prognostic factors, does not result in a substantial improvement in the prediction of poor recovery. Also, no recommendations can be given for preferring either the 4DSQ or the Brief IPQ-DLV to assess psychological factors.

Disconnectomics of the Rich Club Impacts Motor Recovery After Stroke

Background and Purpose: Structural brain networks possess a few hubs, which are not only highly connected to the rest of the brain but are also highly connected to each other. These hubs, which form a rich-club, play a central role in global brain organization. To investigate whether the concept of rich-club sheds new light on poststroke recovery, we applied a novel network-theoretical quantification of lesions to patients with stroke and compared the outcomes with what lesion size alone would indicate. Methods: Whole-brain structural networks of 73 patients with ischemic stroke were reconstructed using diffusion-weighted imaging data. Disconnectomes, a new type of network analyses, were constructed using only those fibers that pass through the lesion. Fugl-Meyer upper extremity scores and their changes were used to determine whether the patients show natural recovery or not. Results: Cluster analysis revealed 3 patient clusters: small-lesion-good-recovery, midsized-lesion-poor-recovery (MLPR), and large-lesion-poor-recovery (LLPR). The small-lesion-good-recovery consisted of subjects whose lesions were small, and whose prospects for recovery were relatively good. To explain the nondifference in recovery between the MLPR and LLPR clusters despite the difference (LLPR>MLPR) in lesion volume, we defined the metric to be the sum of the entries in the disconnectome and, more importantly, the to be the sum of all entries in the disconnectome corresponding to edges with at least one node in the rich-club. Unlike lesion volume and corticospinal tract damage (MLPR<LLPR), for , this relationship was reversed (MLPR>LLPR) or showed no difference for . Conclusions: Smaller lesions that focus on the rich-club can be just as devastating as much larger lesions that do not focus on the rich-club, pointing to the role of the rich-club as a backbone for functional communication within brain networks and for recovery from stroke.