Prediction of low bone mineral density in postmenopausal women by artificial neural network model compared to logistic regression model.

Measuring bone mineral density (BMD) is currently the best modality to diagnose osteoporosis and predict future fractures. The use of risk factors to predict BMD and fracture risk has been considered to be inadequate for precise diagnostic purpose, but it may be helpful as a screening tool to determine who actually needs BMD assessment. Recently, artificial neural network (ANN), a nonlinear computational model, has been used in clinical diagnosis and classification. In the present study, we evaluated the risk factors associated with low BMD in Thai postmenopausal women and assessed the prediction of low BMD using an ANN model compared to a logistic regression model. The subjects consisted of 129 Thai postmenopausal women divided into 2 groups, 100 subjects in the training set and the remaining 29 subjects in the validation set. The subjects were classified as having either low BMD or normal BMD by using BMD value 1 SD lower than the mean value of young adults as the cutoff point. Decreased body weight, decreased hip circumference and increased years since menopause were found to be associated with low BMD at the lumbar spine by logistic regression. For the femoral neck, increased age and decreased urinary calcium were associated with low BMD. The models had a sensitivity of 85.0 per cent, a specificity of 11.1 per cent and an accuracy of 62.0 per cent for the diagnosis of low BMD at the lumbar spine when tested in the validation group. For the femoral neck, the sensitivity, specificity and accuracy were 90.5 per cent, 12.5 per cent, and 69.0 per cent, respectively. Models based on ANN correctly classified 65.5 per cent of the subjects in the validation group according to BMD at the lumbar spine with a sensitivity of 80.0 per cent and a specificity of 33.3 per cent while it correctly classified 58.6 per cent of the subjects at the femoral neck with a sensitivity of 76.2 per cent and a specificity of 12.5 per cent. There was no significant difference in terms of accuracy, sensitivity and specificity in the prediction of low BMD at the lumbar spine or the femoral neck between ANN model and logistic regression model. We concluded that ANN does not perform better than convention statistical methods in the prediction of low BMD. The less than perfect performance of the prediction rules used in the prediction of low BMD may be due to the lack of adequate association between the commonly used risk factors and BMD rather than the nature of the computational models.

Evaluation of five commercial kits for free thyroxine estimation by comparison with equilibrium dialysis method in nonthyroidal illness.

Five commercial kits for estimating FT4 in serum of 59 euthyroid control and 38 patients with severe NTI were studied: one non analog method (Gammacoat two step RIA, Clinical Assay) and four different analog methods (Amerlex-M RIA, Amersham; Enzymun test competitive enzyme immunoassay, Boehringer Mannheim; Amerlite chemiluminescence immunoassay, Kodak Clinical Diagnostics; Berilux chemiluminescence immunoassay, Behring) compared with equilibrium dialysis (Eq) method. Serum FT4 estimates in NTI patients measured by all commercial kits in this study yielded results comparable with those by equilibrium dialysis. The proportions of serum FT4 values concordant with FT4 (Eq) in each kit were 76.3, 76.3, 76.3, 68.4 and 78.9 per cent respectively. The percentage of NTI patients who had serum FT4 values lower than the reference levels of the methods used were 21.2, 26.3, 7.9, 15.8, 18.4 and 18.4 per cent respectively. No patient with low serum FT4 (Eq) level had subnormal or high serum TSH value. However, 4 out of 6 patients with high serum FT4 (Eq) values had depressed serum TSH values. All of them also had elevation of serum FT4 estimates measured by all kits. Serum FT4 estimates measured by all methods correlated well with FT4 (Eq) levels within the NTI group.

Serum vitamin D, parathyroid hormone and biochemical markers of bone turnover in normal Thai subjects.

The sera from 158 healthy Thai volunteers (77 males and 81 females), aged 20-80 years, were studied. The vitamin D status, parathyroid gland activity and the magnitude of bone turnover were assessed by measurement of serum 25-hydroxycholecalciferol (25-OH-D), intact parathyroid hormone (N-tact-PTH), osteocalcin and alkaline phosphatase. The mean serum 25-OH-D, N-tact-PTH, osteocalcin and alkaline phosphatase concentrations in men were 67.4 +/- 31.6 (S.D.) [Range (R): 20.6-147.1 ng/ml], 23.3 +/- 10.3 (R: 5.6-56.6 pg/ml) 3.4 +/- 1.5 (R: 1.2-10.5 ng/ml), and 19.9 +/- 6.6 (R: 7.5-35.7 IU/L), respectively, and the mean levels in women were 42.4 +/- 23.9 (R: 13.8-127.8 ng/ml), 26.1 +/- 11.3 (R: 10.5-68.7 pg/ml), 3.7 +/- 2.1 (R: 0.5-11.5 ng/ml), and 19.5 +/- 6.0 (R: 9.1-41.5 IU/L), respectively. There is no evidence of vitamin D deficiency in ambulatory elderly Thais. Serum N-tact PTH increased with advancing age in both men and women whereas increasing serum osteocalcin and alkaline phosphatase with age were observed only in women. In addition, serum alkaline phosphatase correlated to serum osteocalcin only in women suggesting an increase in bone turnover after menopause. These basic data would be useful for the study of metabolic bone diseases in Thai population.

Evaluation of a new direct radioimmunoassay for serum thyroid peroxidase and thyroglobulin antibodies.

The measurement of serum TPO Ab and Tg Ab by a new direct sensitive RIA in this study are quantitative and provided a convenient system. When compared to the commonly used PH technique for TM Ab and Tg Ab, this RIA determination appears to be more sensitive than by PH, since it enabled detection of TPO Ab and/or Tg Ab in sera that were negative by PH. Thus, this RIA determination should be more widely used in a clinical laboratory.

The performances of newer free thyroxine assays vs equilibrium dialysis.

The performances of 5 different commercial kits for the measurement of serum FT4 concentration; Amerlex-M Free T4 RIA, Gammacoat 125I FT4 2 step RIA, Enzymune-test FT4, Amerlite FT4, Berilux FT4 were compared with equilibrium dialysis. All assays demonstrated FT4 values in good correlation with those measured by equilibrium dialysis (r = 0.84-0.89, p < 0.001). The intra-assay coefficients of variation were good and comparable. All methods yielded good separation of uncomplicated hypo/hyperthyroid patients from euthyroid controls. However, only equilibrium dialysis gave consistently normal results of FT4 values throughout pregnancy. All methods other than equilibrium dialysis gave falsely low values in majority of women in the 2nd and 3rd trimester of pregnancy.

Screening for congenital hypothyroidism in Thailand: has its time come?

The incidence of congenital hypothyroidism at Ramathibodi Hospital was approximately 1:2,486-1:3,843 livebirths, comparable to those found in other industrialized countries. The screening program utilizing TSH measured by a kit manufactured in Thailand was accurate and cost-benefit. It is time to organize TSH screening at least in a university hospital in Thailand to prevent the social burden of raising mentally-handicapped children.