Dataset Information

Assessing facial weakness in myasthenia gravis with facial recognition software and deep learning.

ABSTRACT:

Objective

Myasthenia gravis (MG) is an autoimmune disease leading to fatigable muscle weakness. Extra-ocular and bulbar muscles are most commonly affected. We aimed to investigate whether facial weakness can be quantified automatically and used for diagnosis and disease monitoring.

Methods

In this cross-sectional study, we analyzed video recordings of 70 MG patients and 69 healthy controls (HC) with two different methods. Facial weakness was first quantified with facial expression recognition software. Subsequently, a deep learning (DL) computer model was trained for the classification of diagnosis and disease severity using multiple cross-validations on videos of 50 patients and 50 controls. Results were validated using unseen videos of 20 MG patients and 19 HC.

Results

Expression of anger (p = 0.026), fear (p = 0.003), and happiness (p < 0.001) was significantly decreased in MG compared to HC. Specific patterns of decreased facial movement were detectable in each emotion. Results of the DL model for diagnosis were as follows: area under the curve (AUC) of the receiver operator curve 0.75 (95% CI 0.65-0.85), sensitivity 0.76, specificity 0.76, and accuracy 76%. For disease severity: AUC 0.75 (95% CI 0.60-0.90), sensitivity 0.93, specificity 0.63, and accuracy 80%. Results of validation, diagnosis: AUC 0.82 (95% CI: 0.67-0.97), sensitivity 1.0, specificity 0.74, and accuracy 87%. For disease severity: AUC 0.88 (95% CI: 0.67-1.0), sensitivity 1.0, specificity 0.86, and accuracy 94%.

Interpretation

Patterns of facial weakness can be detected with facial recognition software. Second, this study delivers a 'proof of concept' for a DL model that can distinguish MG from HC and classifies disease severity.

SUBMITTER: Ruiter AM

PROVIDER: S-EPMC10424649 | biostudies-literature | 2023 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Assessing facial weakness in myasthenia gravis with facial recognition software and deep learning.

Ruiter Annabel M AM Wang Ziqi Z Yin Zhao Z Naber Willemijn C WC Simons Jerrel J Blom Jurre T JT van Gemert Jan C JC Verschuuren Jan J G M JJGM Tannemaat Martijn R MR

Annals of clinical and translational neurology 20230609 8

<h4>Objective</h4>Myasthenia gravis (MG) is an autoimmune disease leading to fatigable muscle weakness. Extra-ocular and bulbar muscles are most commonly affected. We aimed to investigate whether facial weakness can be quantified automatically and used for diagnosis and disease monitoring.<h4>Methods</h4>In this cross-sectional study, we analyzed video recordings of 70 MG patients and 69 healthy controls (HC) with two different methods. Facial weakness was first quantified with facial expression ...[more]

PMID: 37292032

Similar Datasets

Project description:Myasthenia gravis (MG) is an autoimmune disease affecting the neuromuscular junction, whose clinical hallmark is muscle weakness and early fatigability. Azathioprine (AZA) is commonly used in Myasthenia Gravis therapy. AZA is a purine antagonist, which inhibits the cell cycle in the resting and DNA synthesis phases. It is usually used as an immunosuppressant to block T- and B-cell proliferation. AZA, bioconverted to 6-mercaptopurine by glutathione S-transferase (GST), can be metabolized either through the hypoxanthine phosphoribosyl transferase pathway to active 6-thioguanine nucleotides (6-TGN) or through the thiopurine S-methyltransferase (TPMT) pathway to inactive methyl-thiopurine metabolites. The incorporation of active 6-TGNs into DNA, causing breaks in DNA strands resulting in interference with RNA production and thereby protein synthesis, is responsible for the drug effect. The response to AZA is determined by which metabolic pathway is being favored. While balanced use of both HPRT and TPMT pathways results in responsiveness to AZA, hyperactivity of TPMT skews the balance towards the TPMT pathway and results in unresponsiveness to AZA with no pharmacological effect. In contrast, low TPMT activity skews the balance towards the HPRT pathway, resulting in increasing side-effects due to the accumulation of 6-TGNs. Current treatments for MG therapy are often inadequate because less than 40% of patients achieve complete remission with available drugs. This reflects the lack of drugs acting on target sites for which there is strong evidence of pathogenicity and the inability to identify responder patients. No criteria for responsiveness are available, exposing patients to unpredictable failures and unpredictable side effects. These individuals are at particular risk for adverse drug reaction (ADRs) or therapeutic failure. Genetic profiling with the Affymetrix drug metabolizing enzymes and transports genotyping array offers the ability to determine 1,931 variants and 225 genes involved in drug metabolism and disposition. Accordingly, the study of well-known drug-metabolizing genes can be involved in the specific metabolic pathway of a drug, which is more likely to define genotype-phenotype association and thereby genotype profiles relevant to drug response that can be applied as predictive biomarkers for pharmacological treatment.

Dataset Information

Assessing facial weakness in myasthenia gravis with facial recognition software and deep learning.

Objective

Methods

Results

Interpretation

Publications

Assessing facial weakness in myasthenia gravis with facial recognition software and deep learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets