Artificial intelligence-based clustering and characterization of Parkinson's disease trajectories

Birkenbihl, Colin; Ahmad, Ashar; Massat, Nathalie J.; Raschka, Tamara; Avberšek, Andreja; Downey, Patrick; Armstrong, Martin; Fröhlich, Holger

doi:10.1038/s41598-023-30038-8

Cited by 12 publications

(6 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the past, data-driven clustering of multisymptom disease trajectories has shown promising results in other neurodegenerative disorders, such as Alzheimer’s or Parkinson’s [ 28 , 30 , 31 ]. The rate of progression is typically variable across the disease trajectory in HD, with the steepest decline in function being seen in early to mid-disease stages [ 32 ].…”

Section: Introductionmentioning

confidence: 99%

Unraveling progression subtypes in people with Huntington’s disease

Raschka,

Li,

Gaßner

et al. 2024

EPMA Journal

Self Cite

View full text Add to dashboard Cite

Background Huntington’s disease (HD) is a progressive neurodegenerative disease caused by a CAG trinucleotide expansion in the huntingtin gene. The length of the CAG repeat is inversely correlated with disease onset. HD is characterized by hyperkinetic movement disorder, psychiatric symptoms, and cognitive deficits, which greatly impact patient’s quality of life. Despite this clear genetic course, high variability of HD patients’ symptoms can be observed. Current clinical diagnosis of HD solely relies on the presence of motor signs, disregarding the other important aspects of the disease. By incorporating a broader approach that encompasses motor as well as non-motor aspects of HD, predictive, preventive, and personalized (3P) medicine can enhance diagnostic accuracy and improve patient care. Methods Multisymptom disease trajectories of HD patients collected from the Enroll-HD study were first aligned on a common disease timescale to account for heterogeneity in disease symptom onset and diagnosis. Following this, the aligned disease trajectories were clustered using the previously published Variational Deep Embedding with Recurrence (VaDER) algorithm and resulting progression subtypes were clinically characterized. Lastly, an AI/ML model was learned to predict the progression subtype from only first visit data or with data from additional follow-up visits. Results Results demonstrate two distinct subtypes, one large cluster (n = 7122) showing a relative stable disease progression and a second, smaller cluster (n = 411) showing a dramatically more progressive disease trajectory. Clinical characterization of the two subtypes correlates with CAG repeat length, as well as several neurobehavioral, psychiatric, and cognitive scores. In fact, cognitive impairment was found to be the major difference between the two subtypes. Additionally, a prognostic model shows the ability to predict HD subtypes from patients’ first visit only. Conclusion In summary, this study aims towards the paradigm shift from reactive to preventive and personalized medicine by showing that non-motor symptoms are of vital importance for predicting and categorizing each patients’ disease progression pattern, as cognitive decline is oftentimes more reflective of HD progression than its motor aspects. Considering these aspects while counseling and therapy definition will personalize each individuals’ treatment. The ability to provide patients with an objective assessment of their disease progression and thus a perspective for their life with HD is the key to improving their quality of life. By conducting additional analysis on biological data from both subtypes, it is possible to gain a deeper understanding of these subtypes and uncover the underlying biological factors of the disease. This greatly aligns with the goal of shifting towards 3P medicine.

show abstract

Section: Introductionmentioning

confidence: 99%

Unraveling progression subtypes in people with Huntington’s disease

Raschka,

Li,

Gaßner

et al. 2024

EPMA Journal

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

“…9 Other researchers identified subtypes using data-driven methods and machine learning. [10][11][12][13][14] These approaches have the advantages of being hypothesis-free and being able to capture more complex patterns from multivariate data. Subtypes were mostly inferred based on cross-sectional differences 10 , but some researchers also investigated differences in disease progression using longitudinal data from single cohorts.…”

Section: Introductionmentioning

confidence: 99%

“…Subtypes were mostly inferred based on cross-sectional differences 10 , but some researchers also investigated differences in disease progression using longitudinal data from single cohorts. 11,12 Our study aims to identify PD subtypes with a focus on differences in disease progression, inferred from multimodal longitudinal cohort data. We extensively characterized PD subtypes regarding differences in motor and non-motor symptom progression, mortality, treatment response, DaTSCAN imaging and digital gait biomarkers.…”

Section: Introductionmentioning

confidence: 99%

“…AMMERLANN 2 , Giuseppe ARENA 1 , Rudi BALLING 1 , Michele BASSIS 1 , Katy BEAUMONT 2 , Regina BECKER 1 , Camille BELLORA 2 , Guy BERCHEM 3 , Daniela BERG 11 , Alexandre BISDORFF 5 , Ibrahim BOUSSAAD 1 , Kathrin BROCKMANN 11 , Jessica CALMES 2 , Lorieza CASTILLO 2 , Gessica CONTESOTTO 2 , Nico DIEDERICH 3 , Rene DONDELINGER 5 , Daniela ESTEVES 2 , Guy FAGHERAZZI 2 , Jean-Yves FERRAND 2 , Manon GANTENBEIN 2 , Thomas GASSER 11 , Piotr GAWRON 1 , Soumyabrata GHOSH 1 , Marijus GIRAITIS 2 , 3 , Enrico GLAAB 1 , Elisa GÓMEZ DE LOPE 1 , Jérôme GRAAS 2 , Mariella GRAZIANO 17 , Valentin GROUES 1 , Anne GRÜNEWALD 1 , Wei GU 1 , Gaël HAMMOT 2 , Anne-Marie HANFF 2,20,21 , Linda HANSEN 1,3 , Michael HENEKA 1 , Estelle HENRY 2 , Sylvia HERBRINK 6 , Sascha HERZINGER 1 , Michael HEYMANN 2 , Michele HU 8 , Alexander HUNDT 2 , Nadine JACOBY 18 , Jacek JAROSLAW LEBIODA 1 , Yohan JAROSZ 1 , Sonja JÓNSDÓTTIR 2 , Quentin KLOPFENSTEIN 1 , Jochen KLUCKEN 1,2,3 , Rejko KRÜGER 1,2,3 , Pauline LAMBERT 2 , Zied LANDOULSI 1 , Roseline LENTZ 7 , Inga LIEPELT 11 , Robert LISZKA 14 , Laura LONGHINO 3 , Victoria LORENTZ 2 , Paula Cristina LUPU 2 , Tainá M. MARQUES 1 , Clare MACKAY 10 , Walter MAETZLER 15 , Katrin MARCUS 13 , Guilherme MARQUES 2 , Patricia MARTINS CONDE 1 , Patrick MAY 1 , Deborah MCINTYRE 2 , Chouaib MEDIOUNI 2 , Francoise MEISCH 1 , Myriam MENSTER 2 , Maura MINELLI 2 , Michel MITTELBRONN 1,4 , Brit MOLLENHAUER 12 , Friedrich MÜHLSCHLEGEL 4 , Romain NATI 3 , Ulf NEHRBASS 2 , Sarah NICKELS 1 , Beatrice NICOLAI 3 , Jean-Paul NICOLAY 19 , Fozia NOOR 2 , Marek OSTASZEWSKI 1 , Clarissa P. C. GOMES 1 , Sinthuja PACHCHEK 1 , Claire PAULY 1,3 , Laure PAULY 2,20 , Lukas PAVELKA…”

mentioning

confidence: 99%

See 2 more Smart Citations

Progression Subtypes in Parkinson’s Disease: A Data-driven Multi-Cohort Analysis

Hähnel,

Raschka,

Sapienza

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

BackgroundThe progression of Parkinson’s disease (PD) is heterogeneous across patients. This heterogeneity complicates patients counseling and inflates the number of patients needed to test potential neuroprotective treatments. Moreover, disease subtypes might require different therapies. This work uses a data-driven approach to investigate how observed heterogeneity in PD can be explained by the existence of distinct PD progression subtypes.MethodsTo derive stable PD progression subtypes in an unbiased manner, we analyzed multimodal longitudinal data from three large PD cohorts. A latent time joint mixed-effects model (LTJMM) was used to align patients on a common disease timescale. Progression subtypes were identified by variational deep embedding with recurrence (VaDER). These subtypes were then characterized across the three cohorts using clinical scores, DaTSCAN imaging and digital gait biomarkers. To assign patients to progression subtypes from baseline data, we developed predictive models and performed extensive cross-cohort validation.ResultsIn each cohort, we identified a fast-progressing and a slow-progressing subtype. These subtypes were reflected by different patterns of motor and non-motor symptoms progression, survival rates, treatment response and features extracted from DaTSCAN imaging and digital gait assessments. Predictive models achieved robust performance with ROC-AUC up to 0.79 for subtype identification. Simulations demonstrated that enriching clinical trials with fast-progressing patients based on predictions from baseline can reduce the required cohort size by 43%.ConclusionOur results show that heterogeneity in PD can be explained by two distinct subtypes of PD progression that are stable across cohorts and can be predicted from baseline data. These subtypes align with the brain-first vs. body-first concept, which potentially provides a biological explanation for subtype differences. The predictive models will enable clinical trials with significantly lower sample sizes by enriching fast-progressing patients.

show abstract