Search

Automated machine learning is explored to predict post-systolic shortening index which is recognized as a ventricular ischemic marker.

Automated machine learning is explored to develop predictive models for post-SCI AIS score improvement.

Automated machine learning is explored to predict localised failure and overall survival among patients suffering from meningioma.

For secondary analysis of CRASH-3 RCT, automated machine learning is explored to predict in-hospital mortality among acute traumatic brain injury patients randomly assigned to Tranexamic acid management.

Automated machine learning is explored with the incorporation of ensemble stacked approach to predict postoperative delirium after microvascular decompression for primary cranial nerve lesions.

Automated machine learning is adopted to predict the survival of uterine serous carcinoma patients by taking in transcriptomic data.

Automated machine learning is explored to predict the quality of life via cognitive functioning parameters outlined in the Wechsler Intelligence Scale for Children in pediatric brain tumour patients.

Automated machine learning is explored to predict the 2-year mortality for astrocytic brain tumour patients by taking into account proteasomal functional parameters.

Automated machine learning is adopted to predict mortality among hospitalised COVID-19 patients.

Automated machine learning is adopted to predict peripheral neuropathy in Parkinsonism and Parkinson's disease patients.

Automated machine learning is explored to develop risk models predicting surgical site infections and adverse events including seroma formation after complex spinal surgeries.

Automated machine learning is explored to develop a sensitive risk predictor for cerebral infarction in patients presenting with subarachnoid haemorrhage.

Automated machine learning is explored to develop a sensitive risk predictor for postoperative delirium in elderly Parkinson's Disease patients who have received deep brain stimulation surgical intervention.

Automated machine learning is adopted to develop prognostication models for overall survival and progression-free survival after a patient suffering from glioblastoma undertakes adjuvant escalated radiotherapy.

Automated Machine Learning is explored to predict AIHD in patients presenting with atypical chest pain with an ensembled stacked approach incorporated.

The current state-of-the-art for automated machine learning is adopted to predict Alzheimer's disease (AD) by adopting variables such as Mini Mental State Examination score, estimated total intracranial volume and Atlas Scaling Factor. A macro-weighted average Area under the Response-operating Curve of 0.96 is achieved with a close-to-perfect AD detection score after incorporating the ensemble approach. Such predictive models shall serve to optimize risk stratification and management protocols for this enfeebling ailment.

A meta-analysis of meta-analyses to determine the clinical efficacy of Azithromycin vis a vis COVID-19

Automated machine learning is explored to develop survival time predictive models for anaplastic oligodendroglioma by adopted data from the Surveillance, Epidemiology, and End Results (SEER) database. Such models, when incorporated into risk stratification protocols, would optimize the outcomes and translate into the reduction of morbidity and mortality associated with this neoplastic condition.

Ensemble approach is incorporated into Automated machine learning to predict Glasgow outcome scale and In-hospital mortality in patients receiving Barbiturate coma therapy for refractory intracranial hypertension post Brain tumour surgery.

Background HF, the underlying cause of a significant proportion of deaths attributed to cardiovascular diseases with diabetes mellitus a universally recognized risk factor. Studies have attested to the significance of early diagnosis and subsequent initiation of treatment as a predictor of PHM. With this context, the current SOTA among auto ML platforms for classification tasks was explored to develop models that accurately, precisely and instantaneously predict PHM. Methods MLjar, the current SOTA among auto ML platforms for classification tasks, was adopted and EA was incorporated to predict PHM in a cohort of 299 HF patients (35% women). The models developed— both: without and with the FUT— using 6 of the time-tested classification algorithms were compared in terms of MCC, F1, ACC and AUROC. Results Upon training without the FUT (scenario 1), an ensemble of RF and NN algorithms achieved the highest AUROC of 89% while predicting PHM. Upon training with the FUT (scenario 2), an ensemble of RF and Llr algorithms exhibited the highest AUROC of 88%. Upon training only with serum creatinine and ejection fraction variables without FUT (scenario 3), an ensemble of RF and Xgboost algorithms achieved an ACC of 80%. The presented models outperformed those developed by Chicco D & Jurman G in all aspects with the exception of the F1 score in scenario 3. Conclusion Adoption of the SOTA auto ML platform and incorporation of EA significantly improve the accuracy and precision of ML models predicting post-HF survival. Such models have the potential to be incorporated into HF management protocols to achieve optimal and instantaneous risk stratification that would potentially translate into a significant decrease in the associated morbidity and mortality.