Logistic Regression and Elastic Net Regularization for the Diagnosis of Fibromyalgia

Fibromyalgia Clinical Trial

— LEDF  

Official title:

Logistic Regression and Elastic Net Regularization for the Diagnosis of Fibromyalgia: A Quantitative Approach Using B-Mode Ultrasound

Clinical Trial Details — Status: Completed

Administrative data

• NCT number: NCT04088747
• Other study ID #: FibromyalgiaDiagnosis
• Status: Completed
• Start date: September 1, 2018
• Est. completion date: September 6, 2019

Study information

• Verified date: September 2019
• Source: Toronto Rehabilitation Institute
• Contact: n/a
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

This study will utilize ultrasound image texture variables to construct an elastic net regularized, logistic regression model to differentiate between healthy and Fibromyalgia patients. The collected ultrasound data will be from participants who are healthy, and from participants who have Fibromyalgia. The predicted performance accuracy of the diagnostic model will be validated and this will confirm or deny the hypothesis that differentiation between the two cohorts is possible.

Description:

Fibromyalgia (FM) diagnosis remains a challenge for clinicians due to a lack of objective diagnostic tools. One proposed solution is the use of quantitative ultrasound (US) techniques, such as image texture analysis, which has demonstrated discriminatory capabilities with other chronic pain conditions. The investigators propose the use of US image texture variables to construct an elastic net regularized, logistic regression model, for differentiating between the trapezius muscle in the healthy and FM patients. 162 Ultrasound videos of the right and left trapezius muscle were acquired from healthy participants and participants with FM. The videos will then be put through a mutli-step processing pipe including converting them into skeletal muscle regions of interest (ROI). The ROI's will be then filtered by an algorithm utilizing the complex wavelet structural similarity index (CW-SSIM), which removes ROI's that are too similar to one another. Eighty-eight texture variables will be extracted from the ROI's, which will be used in nested cross-validation to construct a logistic regression model with and without elastic net regularization. The generalized performance accuracy of both models will be estimated and confirmed with a final validation on a holdout test set. Depending on the predicted, generalized performance accuracy it will be validated or not by the final, holdout test set (confirming the model construction is accurate). These models should then confirm or deny the hypothesis that a regularized logistic regression model built on ultrasound texture features can accurately differentiate between healthy trapezius muscle and that of patients with FM.

Recruitment information / eligibility

• Status: Completed
• Enrollment: 81
• Est. completion date: September 6, 2019
• Est. primary completion date: September 6, 2019
• Accepts healthy volunteers: Accepts Healthy Volunteers
• Gender: All
• Age group: 20 Years to 65 Years

Eligibility

Inclusion Criteria:

• gender independent; chronic widespread pain, fitting the 2016 FM criteria, absence of myofascial pain syndrome trigger points and between the ages of 20 and 65 years (44.3 ± 13.9 years).

• Healthy asymptomatic volunteers who were age matched (n = 17) with no physical complaints or abnormality on physical examination also participated.

Exclusion Criteria:

• Participants were excluded if they demonstrated clinical evidence of another cause for widespread pain, such as polymyositis, dermatomyositis, endocrine disorders, etc. None of the participants had performed any physical exercise during the two to three days prior to entry into the study.

Related Conditions & MeSH terms

• Fibromyalgia
• Myofascial Pain Syndromes

Intervention

Diagnostic Test:
• Ultrasound Imaging
B-mode ultrasound pictures of the upper Trapezius were collected from both left and right sides.

Locations

Country	Name	City	State
Canada	Toronto Rehabilitation Institute	Toronto	Ontario

Sponsors (1)

Lead Sponsor	Collaborator
Toronto Rehabilitation Institute

Country where clinical trial is conducted

Canada, 

References & Publications (22)

Ablin JN, Wolfe F. A Comparative Evaluation of the 2011 and 2016 Criteria for Fibromyalgia. J Rheumatol. 2017 Aug;44(8):1271-1276. doi: 10.3899/jrheum.170095. Epub 2017 Jun 1. — View Citation

Behr M, Noseworthy M, Kumbhare D. Feasibility of a Support Vector Machine Classifier for Myofascial Pain Syndrome: Diagnostic Case-Control Study. J Ultrasound Med. 2019 Aug;38(8):2119-2132. doi: 10.1002/jum.14909. Epub 2019 Jan 7. — View Citation

Bendtsen L, Nørregaard J, Jensen R, Olesen J. Evidence of qualitatively altered nociception in patients with fibromyalgia. Arthritis Rheum. 1997 Jan;40(1):98-102. — View Citation

Bishop, C. M. Pattern recognition and machine learning. New York, NY: Springer-Verlag: 2006. p. 205-207.

Galloway, M. M. Texture classification using gray level run length. Computer graphics and image processing. 1975;4(2):172-179.

Gittins R, Howard M, Ghodke A, Ives TJ, Chelminski P. The Accuracy of a Fibromyalgia Diagnosis in General Practice. Pain Med. 2018 Mar 1;19(3):491-498. doi: 10.1093/pm/pnx155. — View Citation

Haralick, R. M., & Shanmugam, K. Textural features for image classification. IEEE Transactions on systems, man, and cybernetics. 1973;SMC-3(6):610-621.

Jalalian A, Mashohor SB, Mahmud HR, Saripan MI, Ramli AR, Karasfi B. Computer-aided detection/diagnosis of breast cancer in mammography and ultrasound: a review. Clin Imaging. 2013 May-Jun;37(3):420-6. doi: 10.1016/j.clinimag.2012.09.024. Epub 2012 Nov 13. Review. — View Citation

Kravis MM, Munk PL, McCain GA, Vellet AD, Levin MF. MR imaging of muscle and tender points in fibromyalgia. J Magn Reson Imaging. 1993 Jul-Aug;3(4):669-70. — View Citation

Kumbhare DA, Ahmed S, Behr MG, Noseworthy MD. Quantitative Ultrasound Using Texture Analysis of Myofascial Pain Syndrome in the Trapezius. Crit Rev Biomed Eng. 2018;46(1):1-31. doi: 10.1615/CritRevBiomedEng.2017024947. — View Citation

MathWorks. Image Processing Toolbox., Release 2018a, The MathWorks Inc.,Natick, Massachusetts, United States

MathWorks. Statistics and Machine Learning Toolbox., Release 2018a, The MathWorks Inc.,Natick, Massachusetts, United States

Meenagh G, Sakellariou G, Iagnocco A, Delle Sedie A, Riente L, Filippucci E, Di Geso L, Grassi W, Bombardieri S, Valesini G, Montecucco C. Ultrasound imaging for the rheumatologist XXXIX. Sonographic assessment of the hip in fibromyalgia patients. Clin Exp Rheumatol. 2012 May-Jun;30(3):319-21. Epub 2012 Jun 26. — View Citation

Sampat MP, Wang Z, Gupta S, Bovik AC, Markey MK. Complex wavelet structural similarity: a new image similarity index. IEEE Trans Image Process. 2009 Nov;18(11):2385-401. doi: 10.1109/TIP.2009.2025923. Epub 2009 Jun 23. — View Citation

Sarle, W. S. Stopped training and other remedies for overfitting. Computing science and statistics, 1996:352-360.

Schaefer C, Mann R, Masters ET, Cappelleri JC, Daniel SR, Zlateva G, McElroy HJ, Chandran AB, Adams EH, Assaf AR, McNett M, Mease P, Silverman S, Staud R. The Comparative Burden of Chronic Widespread Pain and Fibromyalgia in the United States. Pain Pract. 2016 Jun;16(5):565-79. doi: 10.1111/papr.12302. Epub 2015 May 16. — View Citation

U.S. Department of Health and Human Services Food and Drug Administration/Centre for Drug Evaluation and Research. Guidance for Industry and FDA Staff Qualification Process for Drug Development Tools. Silver Spring, MD: Author; 2014

Virmani, J., Kumar, V., Kalra, N., & Khandelwal, N. Prediction of liver cirrhosis based on multiresolution texture descriptors from B-mode ultrasound. International Journal of Convergence Computing 2013;1(1):19-37.

Wolfe F, Clauw DJ, Fitzcharles MA, Goldenberg DL, Häuser W, Katz RL, Mease PJ, Russell AS, Russell IJ, Walitt B. 2016 Revisions to the 2010/2011 fibromyalgia diagnostic criteria. Semin Arthritis Rheum. 2016 Dec;46(3):319-329. doi: 10.1016/j.semarthrit.2016.08.012. Epub 2016 Aug 30. — View Citation

Wolfe F, Ross K, Anderson J, Russell IJ, Hebert L. The prevalence and characteristics of fibromyalgia in the general population. Arthritis Rheum. 1995 Jan;38(1):19-28. — View Citation

Xian, G. M. An identification method of malignant and benign liver tumors from ultrasonography based on GLCM texture features and fuzzy SVM. Expert Systems with Applications 2010;37(10):6737-6741.

Zou, H., & Hastie, T. Regularization and variable selection via the elastic net. Journal of the royal statistical society: series B (statistical methodology) 2005;67(2):301-320.

* Note: There are 22 references in all — Click here to view all references

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Ultrasound Image Texture Variables	91 statistical image texture variables are extracted from the B mode ultrasound images from both cohorts in order to construct a diagnostic model. The texture variables will be extracted using MATLAB.	1 hour
Primary	Fibromyalgia Diagnostic Criteria	This evaluates symptoms related to Fibromyalgia and determines a score to assess the severity. This score is comprised of the Widespread Pain Index(WPI), which quantifies the regions of pain, and the Symptom Severity Scale(SSS), which measures qualitative aspects of pain such as fatigue and cognitive symptoms. The WPI scale ranges from 0-19 (0- no areas of body pain, 19- all body regions have pain), whereas the SSS ranges from 0-12 (0-no qualitative aspects of pain, 12-many qualitative aspects of pain). This criteria was evaluated on each patient to determine which cohort they belong to. According to the Fibromyalgia Diagnostic Criteria, one is diagnosed with Fibromyalgia if they have a WPI score of 7 or higher, and a SSS score of 5 or higher. Fibromyalgia is also diagnosed with a score of 3-6 on the WPI score, and a score of 9 or higher on the SSS score.	10 minutes
Primary	Central Sensitization Inventory	This is a self reported outcome measure designed to identify patients that experience central sensitization. It involves 25 questions which include symptomatic experiences. The subject must answer on a scale of 0(never) to 5(always) corresponding to how often they experience these. The maximum score is 100 and a score of more than 40 indicates the presence of Central Sensitization. This criteria was evaluated on each patient to determine which cohort they belong to.	10 minutes