Developing an Explainable Machine Learning-Based Thyroid Disease Prediction Model

Healthcare and medicine are key areas where machine learning algorithms are widely used. The medical decision support systems thus created are accurate enough, however, they suffer from the lack of transparency in decision making and shows a black box behavior. However, transparency and trust are significant in the field of health and medicine and hence, a black box system is sub optimal in terms of widespread applicability and reach. Hence, the explainablility of the research make the system reliable and understandable, thereby enhancing its social acceptability. The presented work explores a thyroid disease diagnosis system. SHAP, a popular method based on coalition game theory is used for interpretability of results. The work explains the system behavior both locally and globally and shows how machine leaning can be used to ascertain the causality of the disease and support doctors to suggest the most effective treatment of the disease. The work not only demonstrates the results of machine learning algorithms but also explains related feature importance and model insights.

Analysis of clinical features of myasthenia gravis complicated with hyperthyroidism

Objectives: To investigate the clinical features of patients with myasthenia gravis complicated with and without hyperthyroidism. Methods: A total of 2083 patients with myasthenia gravis (MG) admitted in Center of Treatment of Myasthenia Gravis Hebei Province between January 2013 and July 2020 were retrospectively analyzed and divided into two groups: Group-A and Group-B, with 108 MG patients complicated with hyperthyroidism in Group-A and 1975 MG patients without thyroid disease in Group-B. The age of onset, gender, Osserman classification, acetylcholine receptor antibody and thymus status of the two groups were analyzed in the two groups. Independent-sample t test was used for intra-group comparison, and χ2 test was utilized for comparison of enumeration data. P<0.05 indicates a statistically significant difference. Results: The age of onset in Group-A was significantly lower than that in Group-B (p=0.000), the number of female patients was significantly higher than that in Group-B (p=0.037), and the level of Achrabs titer was significantly lower than that in Group-B (p=0.000). The incidence of thymoma in Group-A was significantly lower than that in Group-B (p=0.012), while the incidence of thymic hyperplasia was significantly higher than that in Group-B (p=0.000). Conclusion: Patients with MG complicated with hyperthyroidism are mainly female, with a lower age of onset, a lower level of acetylcholine receptor antibody, a lower incidence of thymoma, and a higher incidence of thymic hyperplasia. The clinical features of such patients are remarkably different from those of MG without thyroid disease. doi: https://doi.org/10.12669/pjms.38.3.4656 How to cite this:Wang Y, Qi G, Yang Y. Analysis of clinical features of myasthenia gravis complicated with hyperthyroidism. Pak J Med Sci. 2022;38(3):---------. doi: https://doi.org/10.12669/pjms.38.3.4656 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Pathophysiology and Clinical Features of Neuropsychiatric Manifestations of Thyroid Disease

Thyroid hormones (TH) have a cardinal role in the development of the central nervous system during embryogenesis and early infancy. However, the TH responsive genes in the developing brain cease to respond to TH in adulthood. Nevertheless, thyroid dysfunction in adults is commonly associated with a host of cognitive and psychiatric problems. Cognitive decline, dysphoria and depression are common manifestations of overt hypothyroidism while hyperthyroidism can cause agitation, acute psychosis and apathy especially in older people. Whereas levothyroxine treatment can reverse dementia in the setting of hypothyroidism, the effect of levothyroxine on depressive symptoms in subjects with subclinical hypothyroidism is controversial. The use of supraphysiologic doses of TH to treat depression refractory to antidepressant remains a viable therapeutic tool with the caveat that excessive doses of thyroid hormone to treat depression may have potentially damaging effects on other organ systems. The present communication describes the pathophysiology of neuropsychiatric manifestations of thyroid disease including changes in neurotransmission, alterations in neuronal or glial cell gene expression, blood-brain barrier dysfunction, increased risk of cerebrovascular disease and occasionally cerebral inflammatory disease in the context of autoimmune thyroid disease. Elucidating the molecular mechanisms of TH effect on cerebral tissue will help identify novel therapeutic targets for managing people with neuropsychiatric disorders.

Practical Guidelines for Diagnosing and Treating Thyroid Disease Based on the WOMED Metabolic Model of Disease Focusing on Glycolysis and Coenzyme Q10 Deficiency—A Clinical Alternative to the 2021 Retired Clinical Practice Guidelines of the Endocrine Society

This review aims to provide a functional, metabolic view of the pathogenesis of benign thyroid disease. Here, we summarize the features of our previous publications on the “WOMED model of benign thyroid disease”. As of 2021, the current state of art indicates that the basic alteration in benign thyroid disease is a metabolic switch to glycolysis, which can be recognized using 3D-power Doppler ultrasound. A specific perfusion pattern showing enlarged vessels can be found using this technology. This switch originates from an altered function of Complex I due to acquired coenzyme Q10 deficiency, which leads to a glycolytic state of metabolism together with increased angiogenesis. Implementing a combined supplementation strategy that includes magnesium, selenium, and CoQ10, the morphological and perfusion changes of the thyroid can be reverted, i.e., the metabolic state returns to oxidative phosphorylation. Normalization of iron levels when ferritin is lower than 50 ng/mL is also imperative. We propose that a modern investigation of probable thyroid disease requires the use of 3D-power Doppler sonography to recognize the true metabolic situation of the gland. Blood levels of magnesium, selenium, CoQ10, and ferritin should be monitored. Thyroid function tests are complementary so that hypo- or hyperthyroidism can be recognized. Single TSH determinations do not reflect the glycolytic state.

Relative Frequency of Islet Autoimmunity in Children and Adolescents with Autoimmune Thyroid Disease

The present study aims to investigate islet autoimmunity and susceptibility to type 1 diabetes (T1D) in children/adolescents with autoimmune thyroid disease (AITD), and family members of AITD patients with islet autoimmunity. Islet-cell cytoplasmic, glutamic-acid decarboxylase and tyrosine-phosphatase autoantibodies were measured in 161 AITD patients [(127 with autoimmune thyroiditis (AT); 34 with Graves’ disease (GD)], 20 family members of AITD patients with islet autoimmunity, and 155 age-matched controls. Islet autoimmunity was found in 10.6% of AITD patients, significantly more frequent than in controls (1.9%; p=0.002). Higher prevalence of islet autoantibodies was found in females with AITD (p=0.011) but not in males (p=0.16) as well as in AT (p=0.013) but not GD patients (p=0.19), compared to corresponding controls. Two or three islet autoantibodies were found concurrently in six AITD patients with islet autoimmunity. They all developed T1D and had significantly higher islet autoantibodies titers (p=0.01) compared to AITD patients with single islet autoantibodies but normal glucose metabolism. T1D was found in 3.7% of AITD patients compared to 0.2% in age-matched, general Croatian population. Islet autoantibodies were found in 5/20 family members of AITD patients with islet autoimmunity, among which two developed T1D. None of the controls was positive to more than one islet autoantibodies or developed T1D. Conclusion: Children/adolescents with AITD (particularly females and patients with AT) represent a risk group for islet autoimmunity and T1D, as well as family members of AITD patients with positive islet autoantibodies, but last observation must be examined in a larger number of patients.

Autoimmune thyroid disease in diabetic children and adolescents: a single center study from south Punjab

Objective: To evaluate the prevalence of autoimmune thyroid disease (AITD) in diabetic children in south Punjab. Methods: This was an observational cross sectional study from Jan 2019 to Dec 2019 in the outpatient diabetic clinic of the department of pediatric endocrinology at Children Hospital and The Institute of Child Health Multan. A total of 161 consecutive patients of both genders with TIDM were enrolled in this study after taking informed consent. Blood samples for Thyroid functions testes including thyroid stimulating hormone (TSH), free thyroxin (fT4), Thyroid peroxidase antibody (TPO-Ab), thyroglobulin antibody (TG-Ab) and glycosylated hemoglobin (HbA1C) level were sent. Results: Among diabetic children males were 83 (51.6%). Age range was 2-15 years. Mean age and standard deviation was 9.7± 4.3. TPO-Ab was positive in 34 patients (21.1%) and TG-Ab in 27 patients (16.7%), whereas both antibodies were positive in 17 patients (10.5%). Six patients (3.7%) had evidence of subclinical hypothyroidism, 8 patients (4.9%) had overt hypothyroidism and 1 patient (0.62%) had hyperthyroidism Conclusion: The prevalence of AITD among children and adolescents with type 1 diabetes mellitus was 21.1% in our study. Hypothyroidism was more prevalent in these children compared to hyperthyroidism. All diabetic children should be screened for AITD. Thyroid functions should be checked where TPO antibody is positive. Keywords: Autoimmune thyroid disease, anti thyroid peroxidase antibody, anti thyroglobulin Continuous...