Radiomic-based nonlinear supervised learning classifiers on non-contrast CT to predict functional prognosis in patients with spontaneous intracerebral hematoma

PURPOSE

To evaluate if nonlinear supervised learning classifiers based on non-contrast CT can predict functional prognosis at discharge in patients with spontaneous intracerebral hematoma.

METHODS

Retrospective, single-center, observational analysis of patients with a diagnosis of spontaneous intracerebral hematoma confirmed by non-contrast CT between January 2016 and April 2018. Patients with HIE > 18 years and with TCCSC performed within the first 24 h of symptom onset were included. Patients with secondary spontaneous intracerebral hematoma and in whom radiomic variables were not available were excluded. Clinical, demographic and admission variables were collected. Patients were classified according to the Modified Rankin Scale (mRS) at discharge into good (mRS 0-2) and poor prognosis (mRS 3-6). After manual segmentation of each spontaneous intracerebral hematoma, the radiomics variables were obtained. The sample was divided into a training and testing cohort and a validation cohort (70-30% respectively). Different methods of variable selection and dimensionality reduction were used, and different algorithms were used for model construction. Stratified 10-fold cross-validation were performed on the training and testing cohort and the mean area under the curve (AUC) were calculated. Once the models were trained, the sensitivity of each was calculated to predict functional prognosis at discharge in the validation cohort.

RESULTS

105 patients with spontaneous intracerebral hematoma were analyzed. 105 radiomic variables were evaluated for each patient. P-SVM, KNN-E and RF-10 algorithms, in combination with the ANOVA variable selection method, were the best performing classifiers in the training and testing cohort (AUC 0.798, 0.752 and 0.742 respectively). The predictions of these models, in the validation cohort, had a sensitivity of 0.897 (0.778-1;95%CI), with a false-negative rate of 0% for predicting poor functional prognosis at discharge.

CONCLUSION

The use of radiomics-based nonlinear supervised learning classifiers are a promising diagnostic tool for predicting functional outcome at discharge in HIE patients, with a low false negative rate, although larger and balanced samples are still needed to develop and improve their performance.

The diagnostic and prognostic performance of Lp-PLA2 in acute ischemic stroke

OBJECTIVE

We aimed to investigate the correlation between serum lipoprotein-associated phospholipase A2 (Lp-PLA2) level and acute ischemic stroke (AIS) severity and recurrence, which could indicate the diagnostic and prognostic values of Lp-PLA2.

METHODS

Two hundred and fifty-one AIS patients who were diagnosed in the department of neurology, China-Japan Union Hospital, from April 2018 to June 2019 and 100 non-cerebrovascular disease patients were included in this study. Demographic data and clinical materials including age, sex, BMI, medical history, bad habits, imaging data, blood tests, etc., were collected. Stroke severity and risk were evaluated, respectively. AIS patients were followed up for 6 months for stroke recurrence monitoring.

RESULTS

The AIS group had significantly higher Lp-PLA2 level than the control group. High Lp-PLA2 level was the independent risk factor of AIS (OR 1.010; 95% CI 1.007-1.013, P<0.001). Admission NIHSS was compared between Lp-PLA2 categories dichotomized by median. Serum Lp-PLA2 level was positively correlated with NIHSS (r=0.268, P<0.001). Mild stroke group and severe stroke group were defined based on NIHSS. Compared with the mild stroke group, the severe stroke group had higher smoking ratio, posterior circulation stenosis incidence, LDL level, Lp-PLA2 level and admission Essen score. Logistic regression showed Lp-PLA2 levels per 50ng/mL (OR 1.381, 95% CI 1.212-1.573, P<0.001) and Lp-PLA2 category (OR 2.420, 95% CI 1.363-4.297, P=0.003) were both independently associated with severe stroke. During the 6-month follow-up of all 251 AIS patients, 31 stroke recurrence cases were observed. Both Lp-PLA2 levels per 50ng/mL (OR 1.420, 95% CI 1.212-1.664, P<0.001) and Lp-PLA2 category (OR 2.726, 95% CI 1.201-6.178, P=0.016) were significantly associated with the stroke recurrence. Under the receiver Operating Characteristic curve, Lp-PLA2 showed significant predicting ability for stroke recurrence (AUC 0.723; 95% CI 0.620-0.826, P<0.001).

CONCLUSION

High serum Lp-PLA2 level is correlated with AIS incidence, disease severity and recurrence, which could be utilized to guide clinical practice.

[Post-stroke apathy]

Apathy is a motivational disturbance that can be defined as a quantitative reduction of goal-directed behaviour. Patients present with loss of motivation, concern, interest, and emotional response, resulting in a loss of initiative, decreased interaction with their environment, and a reduced interest in social life. Apathy not only appears to be common in stroke patients, but it has also been related to a wide range of negative consequences for the patients and their caregivers, including poor functional recovery, loss of social independence, and caregiver distress. Clear definition and consensus diagnostic criteria for apathy are needed to accomplish an accurate assessment and an individualised treatment plan. Although there have been reports of successful behavioural therapy treatment of apathetic states, there is a paucity of controlled clinical trials on the efficacy of apathetic behaviours using pharmacotherapy.