An informative probability model enhancing real time echobiometry to improve fetal weight estimation accuracy

Neuro-fuzzy classification of asthma and chronic obstructive pulmonary disease

Background

This paper presents a system for classification of asthma and chronic obstructive pulmonary disease (COPD) based on fuzzy rules and the trained neural network.

Methods

Fuzzy rules and neural network parameters are defined according to Global Initiative for Asthma (GINA) and Global Initiative for chronic Obstructive Lung Disease (GOLD) guidelines. For neural network training more than one thousand medical reports obtained from database of the company CareFusion were used. Afterwards the system was validated on 455 patients by physicians from the Clinical Centre University of Sarajevo.

Results

Out of 170 patients with asthma, 99.41% of patients were correctly classified. In addition, 99.19% of the 248 COPD patients were correctly classified. The system was 100% successful on 37 patients with normal lung function. Sensitivity of 99.28% and specificity of 100% in asthma and COPD classification were obtained.

Conclusion

Our neuro-fuzzy system for classification of asthma and COPD uses a combination of spirometry and Impulse Oscillometry System (IOS) test results, which in the very beginning enables more accurate classification.

Additionally, using bronchodilatation and bronhoprovocation tests we get a complete patient's dynamic assessment, as opposed to the solution that provides a static assessment of the patient.

Statistical limits in sonographic estimation of birth weight

PURPOSE

The accuracy of sonographic estimation of birth weight (EBW) is compromised by the precision of the biometrical measurements and the quality of the algorithms. This prospective study was to evaluate technical aspects to derive new equations for the EBW.

METHODS

Three consecutive phases were carried out (1) to recruit a homogenous population, (2) to derive eight new algorithms using a multiple stepwise mathematical/statistical method, and (3) to test the accuracy of the developed equations. Only those patients with a singleton pregnancy who delivered within 48 h from the scan were considered for the analysis.

RESULTS

The study population was made of 473 women. Four polynomial, two square root and two logarithmic algorithms were derived from a balanced study group of 200 women selected from the original study population. These formulas were subsequently applied and compared between them and showed a significant correlation with birth weight (p < 0.0001) and satisfactory statistical performances (r > 0.9), nevertheless they performed similarly to other equations previously published.

CONCLUSIONS

The present findings define better the limitations associated with the intrinsic properties of algorithms and highlight that the possibility to improve the precision of sonographic measurements remains the only point at issue to increase the accuracy in the prediction of birth weight.

Discordant twins: diagnosis, evaluation and management

Approximately 16% of twin gestations have discordance of at least 20%. We identified 14 risk factors for divergent growth that can be categorized as maternal, fetal, or placental. Determination of chorionicity and serial ultrasound evaluation with a high index of suspicion for divergent growth is required for the diagnosis and stratification of risk. The highest reported likelihood ratio for detection of discordance was 5.9 during the first trimester examination and 6.0 for the second trimester. Although our ability to identify discordant twins is limited, once suspected and at viable gestational age, these pregnancies should have antepartum testing. Discordant growth alone is not an indication for preterm birth. Although there are multiple publications on the increased morbidity and mortality rates with discordant growth, there is a paucity of reports on how to manage them optimally and deliver them in a timely manner.