Short abstract
This commentary is on the original article by Greve et al. on pages 100‐106 of this issue.
Standardization of assessment of functioning is a pivotal aspect of improving outcomes in rehabilitation. 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. 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. As CP is a complex neurodevelopmental disorder with a wide range of functional limitations leading to restrictions in activities and participation, the care of children with CP should be managed holistically on all domains of functioning. 2
In rehabilitating children with developmental and neuromotor functional limitations, the International Classification of Functioning, Disability and Health (ICF) is already often used in care practice. NLP can be used in large sets of unstructured text data for the automated assessment of functioning based on multiple categories and levels of the ICF. 3 Rehabilitation of children with developmental and neuromotor conditions might improve by the automated assessments of functioning in ICF categories and levels. Deployment of appropriate and personalized rehabilitation plans for these children requires insight into how their functioning develops over time, so that relevant patterns and predictors can be associated. NLP has a great potential to mine the free text in patient records for significant patterns and insights.
Over the last few years, artifical intelligence (AI)‐based techniques, such as NLP, have increasingly been applied in the health care domain, and this will grow even more in the future. However, to apply AI‐based data analysis in rehabilitation care for children with developmental conditions, ethical considerations will have to be accounted for.
First, one should reflect on the role of various stakeholders, children, parents, and health care professionals, regarding choices in developing and using AI techniques. Which ICF domains are relevant for assessing functioning and improving rehabilitation care? The perspectives of children might differ from those of parents and professionals. It is crucial to take seriously the views and experiences of children, parents, and professionals, and organize dialogues between these groups, to come to well‐founded decisions on which ICF domains are appropriate for rehabilitation care.
Second, we propose combining the ICF with the capability approach. In the capability approach, quality of life is defined as the freedom people have to choose the life they have reason to value. 4 This implies that we should not only focus on actual functioning, but more importantly on providing opportunities for persons to decide how they wish to function in a specific way. AI techniques should be applied so that they do not determine choices, but provide options for choice. Therefore a new view on AI is required; not as a replacement of human intelligence, but as a way to enlarge it. 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. Applying AI techniques should enhance the freedom of children to live the life they value. An ethical framework based on the perspectives of children, parents, professionals, and researchers on the use of AI may be a necessary first step in guiding future AI research, not only for children with developmental and neuromotor conditions but for all patient populations.
DATA AVAILABILITY STATEMENT
Not required
REFERENCES
- 1. Greve K, Ni Y, Bailes AF, et al. Gross motor function prediction using natural language processing in cerebral palsy. Dev Med Child Neurol 2022. Jun 5; 1–7. doi: 10.1111/dmcn.15301. Online ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Rosenbaum PL, Dan B. The continuing evolution of "Cerebral Palsy". Ann Phys Rehabil Med 2020; 63: 387–8. [DOI] [PubMed] [Google Scholar]
- 3. Meskers C, van der Veen S, Kim S, et al. Automated recognition of functioning, activity and participation in COVID‐19 from electronic patient records by natural language processing: a proof‐of‐concept. Ann Med 2022; 54: 235–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Van der Veen S, Evans N, Huisman M, Welch Saleeby P, Widdershoven G. Towards a paradigm shift in healthcare: using the International Classification of Functioning, Disability and Health (ICF) and the capability approach (CA) jointly in theory and practice. Disabil Rehabil 2022. Jun 22; 1–8. doi: 10.1080/09638288.2022.2089737 [DOI] [PubMed] [Google Scholar]
- 5. Hassani H, Silva ES, Unger S, TajMaziani M, Mac Feeley S. Artificial intelligence (AI) or intelligence augmentation (IA): what is the future? AI 2020; 1: 143–55. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Not required