Detecting Parkinson’s disease and its cognitive phenotypes via automated semantic analyses of action stories
Abstract Action-concept outcomes are useful targets to identify Parkinson’s disease (PD) patients and differentiate between those with and without mild cognitive impairment (PD-MCI, PD-nMCI). Yet, most approaches employ burdensome examiner-dependent tasks, limiting their utility. We introduce a fram...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-11-01
|
| Series: | npj Parkinson's Disease |
| Online Access: | https://doi.org/10.1038/s41531-022-00422-8 |
| _version_ | 1797642040700305408 |
|---|---|
| author | Adolfo M. García Daniel Escobar-Grisales Juan Camilo Vásquez Correa Yamile Bocanegra Leonardo Moreno Jairo Carmona Juan Rafael Orozco-Arroyave |
| author_facet | Adolfo M. García Daniel Escobar-Grisales Juan Camilo Vásquez Correa Yamile Bocanegra Leonardo Moreno Jairo Carmona Juan Rafael Orozco-Arroyave |
| author_sort | Adolfo M. García |
| collection | DOAJ |
| description | Abstract Action-concept outcomes are useful targets to identify Parkinson’s disease (PD) patients and differentiate between those with and without mild cognitive impairment (PD-MCI, PD-nMCI). Yet, most approaches employ burdensome examiner-dependent tasks, limiting their utility. We introduce a framework capturing action-concept markers automatically in natural speech. Patients from both subgroups and controls retold an action-laden and a non-action-laden text (AT, nAT). In each retelling, we weighed action and non-action concepts through our automated Proximity-to-Reference-Semantic-Field (P-RSF) metric, for analysis via ANCOVAs (controlling for cognitive dysfunction) and support vector machines. Patients were differentiated from controls based on AT (but not nAT) P-RSF scores. The same occurred in PD-nMCI patients. Conversely, PD-MCI patients exhibited reduced P-RSF scores for both texts. Direct discrimination between patient subgroups was not systematic, but it yielded best outcomes via AT scores. Our approach outperformed classifiers based on corpus-derived embeddings. This framework opens scalable avenues to support PD diagnosis and phenotyping. |
| first_indexed | 2024-03-11T13:54:27Z |
| format | Article |
| id | doaj.art-a331e58c2c4149169ca8627eaf5eb37e |
| institution | Directory Open Access Journal |
| issn | 2373-8057 |
| language | English |
| last_indexed | 2024-03-11T13:54:27Z |
| publishDate | 2022-11-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Parkinson's Disease |
| spelling | doaj.art-a331e58c2c4149169ca8627eaf5eb37e2023-11-02T07:22:40ZengNature Portfolionpj Parkinson's Disease2373-80572022-11-018111010.1038/s41531-022-00422-8Detecting Parkinson’s disease and its cognitive phenotypes via automated semantic analyses of action storiesAdolfo M. García0Daniel Escobar-Grisales1Juan Camilo Vásquez Correa2Yamile Bocanegra3Leonardo Moreno4Jairo Carmona5Juan Rafael Orozco-Arroyave6Global Brain Health Institute, University of CaliforniaGITA Lab, Faculty of Engineering, Universidad de Antioquia UdeAFundación Vicomtech, Basque Research and Technology Alliance (BRTA)Grupo de Neurociencias de Antioquia, Facultad de Medicina, Universidad de AntioquiaSección de Neurología, Hospital Pablo Tobón UribeGrupo Neuropsicología y Conducta (GRUNECO), Facultad de Medicina, Universidad de AntioquiaGITA Lab, Faculty of Engineering, Universidad de Antioquia UdeAAbstract Action-concept outcomes are useful targets to identify Parkinson’s disease (PD) patients and differentiate between those with and without mild cognitive impairment (PD-MCI, PD-nMCI). Yet, most approaches employ burdensome examiner-dependent tasks, limiting their utility. We introduce a framework capturing action-concept markers automatically in natural speech. Patients from both subgroups and controls retold an action-laden and a non-action-laden text (AT, nAT). In each retelling, we weighed action and non-action concepts through our automated Proximity-to-Reference-Semantic-Field (P-RSF) metric, for analysis via ANCOVAs (controlling for cognitive dysfunction) and support vector machines. Patients were differentiated from controls based on AT (but not nAT) P-RSF scores. The same occurred in PD-nMCI patients. Conversely, PD-MCI patients exhibited reduced P-RSF scores for both texts. Direct discrimination between patient subgroups was not systematic, but it yielded best outcomes via AT scores. Our approach outperformed classifiers based on corpus-derived embeddings. This framework opens scalable avenues to support PD diagnosis and phenotyping.https://doi.org/10.1038/s41531-022-00422-8 |
| spellingShingle | Adolfo M. García Daniel Escobar-Grisales Juan Camilo Vásquez Correa Yamile Bocanegra Leonardo Moreno Jairo Carmona Juan Rafael Orozco-Arroyave Detecting Parkinson’s disease and its cognitive phenotypes via automated semantic analyses of action stories npj Parkinson's Disease |
| title | Detecting Parkinson’s disease and its cognitive phenotypes via automated semantic analyses of action stories |
| title_full | Detecting Parkinson’s disease and its cognitive phenotypes via automated semantic analyses of action stories |
| title_fullStr | Detecting Parkinson’s disease and its cognitive phenotypes via automated semantic analyses of action stories |
| title_full_unstemmed | Detecting Parkinson’s disease and its cognitive phenotypes via automated semantic analyses of action stories |
| title_short | Detecting Parkinson’s disease and its cognitive phenotypes via automated semantic analyses of action stories |
| title_sort | detecting parkinson s disease and its cognitive phenotypes via automated semantic analyses of action stories |
| url | https://doi.org/10.1038/s41531-022-00422-8 |
| work_keys_str_mv | AT adolfomgarcia detectingparkinsonsdiseaseanditscognitivephenotypesviaautomatedsemanticanalysesofactionstories AT danielescobargrisales detectingparkinsonsdiseaseanditscognitivephenotypesviaautomatedsemanticanalysesofactionstories AT juancamilovasquezcorrea detectingparkinsonsdiseaseanditscognitivephenotypesviaautomatedsemanticanalysesofactionstories AT yamilebocanegra detectingparkinsonsdiseaseanditscognitivephenotypesviaautomatedsemanticanalysesofactionstories AT leonardomoreno detectingparkinsonsdiseaseanditscognitivephenotypesviaautomatedsemanticanalysesofactionstories AT jairocarmona detectingparkinsonsdiseaseanditscognitivephenotypesviaautomatedsemanticanalysesofactionstories AT juanrafaelorozcoarroyave detectingparkinsonsdiseaseanditscognitivephenotypesviaautomatedsemanticanalysesofactionstories |