Work Description
Title: A Comprehensive Patient-Specific Prediction Model for Temporomandibular Joint Osteoarthritis Progression Open Access Deposited
Attribute | Value |
---|---|
Methodology |
|
Description |
|
Creator | |
Depositor |
|
Contact information | |
Discipline | |
Funding agency |
|
Keyword | |
Resource type | |
Last modified |
|
Published |
|
DOI |
|
License |
To Cite this Work:
(2024). A Comprehensive Patient-Specific Prediction Model for Temporomandibular Joint Osteoarthritis Progression [Data set], University of Michigan - Deep Blue Data. https://doi.org/10.7302/xc32-4d53
(2024). A Comprehensive Patient-Specific Prediction Model for Temporomandibular Joint Osteoarthritis Progression [Data set], University of Michigan - Deep Blue Data. https://doi.org/10.7302/xc32-4d53
Relationships
- This work is not a member of any user collections.
Files (Count: 6; Size: 35.9 GB)
Thumbnailthumbnail-column | Title | Original Upload | Last Modified | File Size | Access | Actions |
---|---|---|---|---|---|---|
![]() |
Readme.txt | 2024-01-15 | 2024-01-15 | 7.47 KB | Open Access |
|
![]() |
All_clinical_data__Aim1_Nov_2021.xlsx | 2024-01-10 | 2024-01-10 | 68.8 KB | Open Access |
|
![]() |
Articular_fossa__Quantitiative_T....xlsx | 2024-01-11 | 2024-01-11 | 37.5 KB | Open Access |
|
![]() |
Condyle_Quantitiative_Trabecular....xlsx | 2024-01-11 | 2024-01-11 | 37.3 KB | Open Access |
|
![]() |
Baseline.zip | 2024-01-12 | 2024-01-12 | 17 GB | Open Access |
|
![]() |
Follow_up.zip | 2024-01-12 | 2024-01-12 | 18.9 GB | Open Access |
|
Date: 15 January, 2024
Dataset Title: A Comprehensive Patient-Specific Prediction Model for Temporomandibular Joint Osteoarthritis Progression
Dataset Creators: L.H.S. Cevidanes, N. Al Turkestani, C. T. Mattos
Dataset Contact: Lucia Cevidanes, luciacev@umich.edu
Funding: NIDCR R01DE024550
Key Points:
- We proposed an open-source predictive modeling framework, which combines feature selection, statistical and machine learning methods, allowing prediction of disease progression.
- Lower values of headache, lower back pain, restless sleep, condyle high grey level-GL- run emphasis, articular fossa GL non uniformity and long run low GL emphasis, saliva levels of Osteoprotegerin and Angiogenin, and higher values of the superior joint space, mouth opening, saliva Vascular-endothelium-growth-factor and Matrix-metalloproteinase-7, serum Epithelial-neutrophil-activating-peptide and age indicate increased the probability of recovery for a specific subject.
- Our multidimensional and multisource analytics tool can enhance clinicians decision-making and pinpoint risk predictors for TMJ OA progression.
Research Overview:
Temporomandibular joint osteoarthritis (TMJ OA) is a multifactorial degenerative disease that affects 8-16 % of the global population. It leads to chronic pain, jaw dysfunction and in advanced stages may require joint replacement. To date, no prognostic tool or single biomarker can accurately predict the course of this intricate disease. Identification of patients at risk for severe prognosis is crucial for timely intervention and reducing the need for surgical management. Hence, we prospectively acquired clinical, imaging and biological data from 106 subjects, with 74 followed over 2-3 years. We proposed an open-source predictive modeling framework, called Ensemble via Hierarchical Predictions through Nested cross-validation tool, which combines 18 feature selection, statistical and machine learning methods, allowing prediction of disease progression with accuracy of 0.87, area under ROC curve of 0.72, and F1 score of 0.82. Importantly, using the interpretable SHAP analysis method, we identified the strongest predictors for TMJ OA progression. Lower values of headache, lower back pain, restless sleep, condyle high grey level-GL- run emphasis, articular fossa GL non uniformity and long run low GL emphasis, saliva levels of Osteoprotegerin and Angiogenin, and higher values of the superior joint space, mouth opening, saliva Vascular-endothelium-growth-factor and Matrix-metalloproteinase-7, serum Epithelial-neutrophil-activating-peptide and age indicate increased the probability of recovery for a specific subject. Our multidimensional and multisource analytics tool can enhance clinicians decision-making and pinpoint risk predictors for TMJ OA progression. The EHPN integrates biological and clinical sciences to solve TMJ OA prognosis in a translational infrastructure that will transform temporomandibular disorders practice.
Methodology:
High resolution cone-beam computed tomography (hr-CBCT) scans were acquired for all participants using the 3D Accuitomo 170 machine. The acquisition protocol included 40 ? 40 mm field of view; 90 kVp, 5 mAs, 30.8 sec scanning time and a voxel size of 0.08 mm.
Instrument and/or Software specifications: NA
Files contained here:
- Excel file for All_clinical_data__Aim1_Nov_2021.xlsx
File Overview
This Excel file contains comprehensive clinical data, compiled in November 2021, for a clinical study. The data is stored in a single sheet titled "MFSDA Jonas Clinical Full Data".
Sheet Structure
Sheet Name: MFSDA Jonas Clinical Full Data
Content: The sheet is structured with patient data across multiple columns, each representing different clinical parameters and patient details.
Columns: The sheet consists of 116 columns. Key columns include:
patient ID: Unique identifier for each patient.
group: Group classification of the patient.
gender: Gender of the patient.
age: Age of the patient.
Clinical parameters such as generalHealth, General Oral Health, Pain Face JawTemple, and various specific facial pain and symptoms health metrics.
Data Details
The first few columns provide an overview of the patient's general health, oral health, pain experience, and all other specific TMJ health related measurements standardized in the Diagnostic Criteria for Temporomandibular disorders are listed in the other columns.
Each row represents a unique patient, with comprehensive data spanning general health assessments to specific clinical measurements.
-Two excel files of the radiomic features measurements of volumes of reference in the condyle lateral pole and in the lateral portion of the articular fossa.
*Articular_fossa__Quantitiative_Trabecular_Bone_Changes.xlsx
File Overview
This Excel file presents quantitative data on trabecular bone changes in the articular fossa, an important aspect of bone health. The data is specifically focused on measurements taken at baseline and follow-up health status checks.
Sheet Structure
Sheet Name: AF
Content: The sheet consists of detailed data for each patient, with a focus on trabecular bone changes in the articular fossa.
Columns: There are 43 columns in the sheet, including:
Patient: Identifier for each patient.
Baseline_Health_status and Follow_up_Health_status: Health status of the patient articular fossa radiomics and morphometry at baseline and follow-up.
Metrics related to trabecular bone characteristics such as Af_energy_BL, Af_entropy_BL, Af_correlation_BL, and others radiomics and morphometry measurements, both at baseline (BL) and follow-up (FU).
Data Details
Each row represents a unique patient, with comprehensive data comparing baseline and follow-up measurements.
The data covers various aspects of articular fossa trabecular bone changes, including energy, entropy, correlation, and other specific metrics.
* Condyle_Quantitiative_Trabecular_Bone_Changes.xlsx
File Overview
This Excel file contains a dataset focused on quantitative changes in the trabecular bone within the mandibular condyle region. The data is organized in a single sheet titled "Sheet1" and is intended for use in clinical research.
Sheet Structure
Sheet Name: Sheet1
Content: The sheet provides patient-specific data, capturing the changes in trabecular bone at baseline and during follow-up visits.
Columns: The dataset consists of 43 columns, which include:
Patient: Identifier for each patient.
Baseline_Health_status and Follow-up_Health_status: Health status of the patient at baseline and follow-up.
Various trabecular bone metrics at baseline (C_energy_BL, C_entropy_BL, C_correlation_BL, etc.) and follow-up (C_energy_FU, C_entropy_FU, C_correlation_FU, etc.).
Data Details
Each row in the dataset represents a unique patient, providing a comparison of trabecular bone characteristics between baseline and follow-up measurements.
The metrics cover various aspects of trabecular bone health, such as energy, entropy, correlation, and other specific radiomics and bone morphometry measures.
-The two folders contain each 148 Cone-beam CT scans for right and left condyles of 74 subjects at Baseline and Follow-up imaging. The folders contents are described below:
*Baseline: 74 right and 74 left mirrored images in nrrd format. All files have consistent image orientation. Filename prefix consist of patient number_ right or left side of the Temporomandibular Joint.
*Follow-up: 74 right and 74 left mirrored images in nii.gz format. All files have consistent image orientation. Filename prefix consist of patient number_ right or left side of the Temporomandibular Joint.
Remediation of Harmful Language
The University of Michigan Library aims to describe library materials in a way that respects the people and communities who create, use, and are represented in our collections. Report harmful or offensive language in catalog records, finding aids, or elsewhere in our collections anonymously through our metadata feedback form. More information at Remediation of Harmful Language.