Wednesday, November 18, 2026
Science
No Result
View All Result
  • Login
  • HOME
  • SCIENCE NEWS
  • CONTACT US
  • HOME
  • SCIENCE NEWS
  • CONTACT US
No Result
View All Result
Scienmag
No Result
View All Result
Home Science News Technology and Engineering

Likelihood Contrast Model Recovers Parkinson’s Trajectory Phenotypes From Sparse Data

November 18, 2026
in Technology and Engineering
Diana Fleming
By Diana Fleming Scienmag Editorial Profile - Neurodegenerative Diseases
Reading Time: 3 mins read
0
Likelihood Contrast Model Recovers Parkinson’s Trajectory Phenotypes From Sparse Data

Likelihood Contrast Model Recovers Parkinson's Trajectory Phenotypes From Sparse Data

65
SHARES
587
VIEWS
Share on FacebookShare on Twitter
ADVERTISEMENT

A statistical modeling framework based on multiclass likelihood contrast classification can accurately identify mathematical trajectory phenotypes from sparse and irregular longitudinal clinical records, according to a study evaluating Parkinson’s disease motor assessment data. In clinical research, patient measurements are frequently collected at unaligned intervals and across irregular visits, presenting significant hurdles for standard classification systems that typically depend on data interpolation or extensive feature reduction. To address these operational challenges, researchers designed an interpretable multiclass likelihood contrast framework for classifying unaligned longitudinal trajectories directly rather than compressing individual patient histories into a restricted set of summary features.

The analytical framework, developed by researchers Sangam Pangeni, Md Rejuan Haque, and Fahad Mostafa from The University of New Mexico Gallup Campus, The Ohio State University Wexner Medical Center, and Arizona State University, relies on class-specific mixed-effects models. Under this formulation, the proposed approach fits class-specific models and classifies each held-out subject according to the largest marginal likelihood. The underlying statistical specification incorporates a quadratic mean trajectory across time while allowing for subject-specific random intercepts and random slopes to capture individual baseline differences and distinct rates of trajectory change.

To evaluate how effectively the method performs under challenging data constraints, the research team conducted both numerical simulation experiments and an empirical evaluation using Parkinson’s disease motor-UPDRS data. Within the Parkinson’s disease experiment, dense longitudinal histories were initially utilized to define three mathematical trajectory phenotypes, providing an objective baseline against which the recovery of trajectory patterns could be systematically tested.

To simulate severe real-world data missingness and test trajectory recovery, the investigators deliberately masked approximately ninety percent of the longitudinal observations from the patient histories. This substantial sparsification evaluated whether the multiclass likelihood contrast algorithm could reliably recover an individual’s trajectory phenotype when provided with only sparse, unaligned records.

Model performance was assessed using subject-level cross-validation and compared directly against an extensive suite of established machine learning and statistical classifiers. Specifically, the evaluation benchmarked the proposed framework against linear mixed-likelihood classification, functional k-nearest neighbors, random forest, support vector machine, polynomial-feature classification, and grid-based principal component analysis classifiers. By testing across these diverse alternative modeling approaches, the researchers determined how effectively the likelihood contrast formulation preserved temporal structure compared to conventional feature-based methods.

Despite the substantial ninety percent sparsification of the data, the proposed model achieved an accuracy of 0.8095, a macro-F1 score of 0.8160, a Matthews correlation coefficient of 0.7107, and a macro-area under the receiver operating characteristic curve of 0.9268. Furthermore, the model exhibited consistent classification performance across different data splits, recording a validation macro-F1 score of 0.7980 alongside a held-out test macro-F1 score of 0.8160, indicating stable generalization without overfitting to specific partitions.

These findings demonstrate that multiclass likelihood contrast learning can preserve informative temporal structure in sparse and irregular longitudinal records without reducing individual trajectories to a small number of summary features. However, the authors emphasized that the Parkinson’s disease findings represent the recovery of data-derived mathematical trajectory phenotypes rather than prospective clinical prediction or diagnostic validation. The research team concluded that further evaluation in larger independent cohorts with clinically defined outcomes and multimodal longitudinal biomarkers is needed to establish genuine clinical utility.

Subject of Research: Multiclass likelihood contrast classification of sparse longitudinal clinical trajectories

Article Title: Parkinson's disease phenotype recovery from sparse longitudinal trajectories using multiclass likelihood contrast classification

Article References: Pangeni, S., Haque, M. R., & Mostafa, F. (2026). Parkinson's disease phenotype recovery from sparse longitudinal trajectories using multiclass likelihood contrast classification. Frontiers in Artificial Intelligence, 9, Article 1968393. https://doi.org/10.3389/frai.2026.1968393

Image Credits: AI Generated

DOI: 10.3389/frai.2026.1968393

Keywords: multiclass likelihood contrast, longitudinal trajectories, Parkinson's disease, motor-UPDRS, sparse longitudinal records, trajectory phenotyping, mixed-effects models, quadratic mean trajectory, random intercepts and slopes, subject-level cross-validation, linear mixed-likelihood classification, functional k-nearest neighbors, random forest, support vector machine, polynomial-feature classification, principal component analysis, Matthews correlation coefficient

Cite Scienmag News

Diana Fleming. (November 18, 2026). Likelihood Contrast Model Recovers Parkinson’s Trajectory Phenotypes From Sparse Data. Scienmag. https://scienmag.com/likelihood-contrast-model-recovers-parkinsons-trajectory-phenotypes-from-sparse-data/

Diana Fleming. "Likelihood Contrast Model Recovers Parkinson’s Trajectory Phenotypes From Sparse Data." Scienmag, 18 November 2026, https://scienmag.com/likelihood-contrast-model-recovers-parkinsons-trajectory-phenotypes-from-sparse-data/. Accessed 18 November 2026.

Diana Fleming. "Likelihood Contrast Model Recovers Parkinson’s Trajectory Phenotypes From Sparse Data." Scienmag. November 18, 2026. https://scienmag.com/likelihood-contrast-model-recovers-parkinsons-trajectory-phenotypes-from-sparse-data/

Tags: functional k-nearest neighborslinear mixed-likelihood classificationlongitudinal trajectoriesMatthews Correlation Coefficientmixed-effects modelsmotor-UPDRSmulticlass likelihood contrastParkinson's diseasepolynomial-feature classificationPrincipal Component Analysisquadratic mean trajectoryRandom Forestrandom intercepts and slopessparse longitudinal recordssubject-level cross-validationsupport vector machinetrajectory phenotyping
Share26Tweet16
Previous Post

Time-to-Practice Metric Proposed to Govern Refugee Clinician Integration in England

Next Post

Justice-Involved US Veterans Face Mortality Rates Up to Five Times Higher

Related Posts

Multi-Batch Quantum Scheduling Reduces Execution Costs on IBM Torino Processor
Technology and Engineering

Multi-Batch Quantum Scheduling Reduces Execution Costs on IBM Torino Processor

November 18, 2026
Kinematically Controlled 3D Printing Transforms Polymer Stringing into Suspended Tissue Scaffolds
Technology and Engineering

Kinematically Controlled 3D Printing Transforms Polymer Stringing into Suspended Tissue Scaffolds

November 18, 2026
Reusing YOLO Features Cuts Tracking Time but Increases Vehicle Association Disagreement
Technology and Engineering

Reusing YOLO Features Cuts Tracking Time but Increases Vehicle Association Disagreement

November 18, 2026
Recursive Human-AI Interaction Can Make Changes in Violence Representation Visible
Technology and Engineering

Recursive Human-AI Interaction Can Make Changes in Violence Representation Visible

November 18, 2026
Hybrid Iterative Model Reveals Oblique Fretting Dynamics in Planetary Gear Systems
Technology and Engineering

Hybrid Iterative Model Reveals Oblique Fretting Dynamics in Planetary Gear Systems

November 18, 2026
Zinc Titanate Photocatalyst Drives High-Yield Biodiesel Synthesis Under Ultraviolet Light
Technology and Engineering

Zinc Titanate Photocatalyst Drives High-Yield Biodiesel Synthesis Under Ultraviolet Light

November 18, 2026
Next Post
Justice-Involved US Veterans Face Mortality Rates Up to Five Times Higher

Justice-Involved US Veterans Face Mortality Rates Up to Five Times Higher

  • Mothers who receive childcare support from maternal grandparents show more optimized

    Mothers who receive childcare support from maternal grandparents show more parental warmth, finds NTU Singapore study

    27656 shares
    Share 11059 Tweet 6912
  • University of Seville Breaks 120-Year-Old Mystery, Revises a Key Einstein Concept

    1061 shares
    Share 424 Tweet 265
  • Bee body mass, pathogens and local climate influence heat tolerance

    682 shares
    Share 273 Tweet 171
  • Researchers record first-ever images and data of a shark experiencing a boat strike

    546 shares
    Share 218 Tweet 137
  • Groundbreaking Clinical Trial Reveals Lubiprostone Enhances Kidney Function

    531 shares
    Share 212 Tweet 133
Science

Embark on a thrilling journey of discovery with Scienmag.com—your ultimate source for cutting-edge breakthroughs. Immerse yourself in a world where curiosity knows no limits and tomorrow’s possibilities become today’s reality!

RECENT NEWS

  • Copper and Cadmium Sulfide Composite Removes Doxycycline Residues in Sunlight
  • Auckland Volcanic Field Past Eruptions Emitted Massive Toxic Gas Quantities
  • Elevated IgE and Allergic History Linked to Pediatric Nephrotic Syndrome Relapse
  • Survey Finds 94 Percent of Tested Israeli Dental Clinics Harbor Contaminated Waterlines

Categories

  • Agriculture
  • Anthropology
  • Archaeology
  • Athmospheric
  • Biology
  • Biotechnology
  • Blog
  • Bussines
  • Cancer
  • Chemistry
  • Climate
  • Earth Science
  • Editorial Policy
  • Marine
  • Mathematics
  • Medicine
  • Pediatry
  • Policy
  • Psychology & Psychiatry
  • Science Education
  • Science News
  • Social Science
  • Space
  • Technology and Engineering

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 5,151 other subscribers

© 2026 Scienmag - Science Magazine

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • HOME
  • SCIENCE NEWS
  • CONTACT US

© 2026 Scienmag - Science Magazine

Discover more from Science

Subscribe now to keep reading and get access to the full archive.

Continue reading