Gross motor function prediction using natural language processing in cerebral palsy

Abstract

Aim: To predict ambulatory status and Gross Motor Function Classification System (GMFCS) levels in patients with cerebral palsy (CP) by applying natural language processing (NLP) to electronic health record (EHR) clinical notes.Method: Individuals aged 8 to 26 years with a diagnosis of CP in the EHR between January 2009 and November 2020 (~12 years of data) were included in a crosssectional retrospective cohort of 2483 patients. The cohort was divided into traintest and validation groups. Positive predictive value, sensitivity, specificity, and area under the receiver operating curve (AUC) were calculated for prediction of ambulatory status and GMFCS levels. Results:The median age was 15 years (interquartile range 10-20 years) for the total cohort, with 56% being male and 75% White. The validation group resulted in 70% sensitivity, 88% specificity, 81% positive predictive value, and 0.89 AUC for predicting ambulatory status. NLP applied to the EHR differentiated between GMFCS levels I-II and III (15% sensitivity, 96% specificity, 46% positive predictive value, and 0.71 AUC); and IV and V (81% sensitivity, 51% specificity, 70% positive predictive value, and 0.75 AUC).Interpretation: NLP applied to the EHR demonstrated excellent differentiation between ambulatory and non-ambulatory status, and good differentiation between GMFCS levels I-II and III, and IV and V. Clinical use of NLP may help to individualize functional characterization and management.

Using artificial intelligence‐based technologies to detect clinically relevant changes of gross motor function in children with cerebral palsy

“…An alternative method by Greve et al uses artificial intelligence via natural language processing to measure motor skills in children with CP. 21…”

Artificial intelligence to improve rehabilitation care for children with developmental conditions: Some ethical considerations

“…Greve et al present the results of using NLP to overcome the large variability in assessing functional mobility in the clinical care of children with cerebral palsy (CP). 1 Their findings offer an encouraging prospect for standardizing diagnostic evaluations of functional mobility over time. The proven successful use of NLP for the evaluation of functional mobility provides opportunities for the standardization of other relevant functional domains for children with CP.…”

“…Information on the epidemiology of post-neonatally acquired cerebral palsy (CP) is critically important because, as Waight et al point out in their new analysis of data from Australia's state CP registers, the causes of post-neonatal CP should be largely preventable. 1 Yet many epidemiological studies of CP exclude cases attributed to post-neonatal causes, and even when studies include these cases, the numbers are often too small to allow evaluation of trends over time or of risk factors or health disparities. 2 For example, studies of the prevalence of CP in the United States conducted by the Center for Disease Control and Prevention in selected states and years have found approximately 7% of CP cases to be due to postneonatal causes, but the numbers from these studies are insufficient for stratified or multivariable analyses to confirm whether trends and disparities seen for CP overall apply to the sub-group of cases with a post-neonatal etiology.…”

“…The use of natural language processing (NLP) provides means to standardize and automate the assessment of functioning from large text data sets. Greve et al present the results of using NLP to overcome the large variability in assessing functional mobility in the clinical care of children with cerebral palsy (CP) 1 . Their findings offer an encouraging prospect for standardizing diagnostic evaluations of functional mobility over time.…”

“…5 We agree on the potential of NLP to 'individualize functional characterization in patients with CP'. 1 Future research could expand to other functional domains than mobility relevant to children with CP and can be assessed and standardized with the same techniques. These domains should be chosen based on dialogue with all relevant stakeholders.…”