Bellamy N.,University of Queensland |
Hochberg M.,University of Maryland Baltimore County |
Tubach F.,University Paris Diderot |
Martin-Mola E.,Hospital Universitario La Paz |
And 9 more authors.
Arthritis Care and Research | Year: 2015
Objective The ability to interpret scores from patient-reported outcome measures at the individual patient level depends on the availability of valid, clinically meaningful benchmarks of response and state attainment. The goal was to develop multinational estimates for minimal clinically important improvement (MCII) and patient acceptable symptomatic state (PASS). Methods A multinational sample of patients with osteoarthritis (OA) was evaluated before and 4 weeks after treatment with nonsteroidal antiinflammatory drugs. Patients completed either the Western Ontario and McMaster Osteoarthritis Index (WOMAC) numerical rating scale 3.1 (hip and knee OA) or the Australian/Canadian Index (AUSCAN) numerical rating scale 3.1 (hand OA) before and after treatment. Patients rated the clinical importance of their response to treatment and their satisfaction with the health state achieved, from which multinational MCII and PASS estimates were calculated for both the WOMAC and AUSCAN indices. Results A total of 609 patients from 7 countries participated in the study. MCII and PASS estimates varied slightly by instrument and subscale. Absolute (percentage) change for MCII ranged 6-9 (10% to 17%) for WOMAC and 4-9 (8% to 15%) for AUSCAN. PASS estimates ranged 39-48 for WOMAC and 38-45 for AUSCAN. Some between-country variation was observed in MCII and PASS. Conclusion Preliminary multinational estimates for MCII and PASS have been developed for several countries. Further research is required to evaluate the robustness, temporal consistency, and age- and sex-dependency of the preliminary estimates as well as their generalizability to other countries, languages, cultures, regions, and other condition-specific outcome measures. © 2015, American College of Rheumatology.
Zakrani A.,Mohammed Vth Souissi University |
Idri A.,Mohammed Vth Souissi University
International Review on Computers and Software | Year: 2010
An important research issue in software project management is how to predict accurately an effort for the software project to develop at an early stage. Achieving high accuracy when estimating the effort of the software under development would supply effectively the project managers in leading to projects finished on time and within budget. Unfortunately, existing software effort estimation techniques are still unable to provide acceptable estimates. This paper investigates the application of Radial Basis Function Neural Networks (RBFN) based on fuzzy clustering in web development effort estimation. The proposed model is designed by integrating the principles of RBFN and the fuzzy C-means clustering algorithm. The architecture of the network is suitably modified at the hidden layer to realize a novel neural implementation of the fuzzy clustering algorithm. Fuzzy set-theoretic concepts are incorporated at the hidden layer, enabling the model to handle uncertain and imprecise data, which can significantly improve the accuracy of the obtained estimates. MMRE and Pred are used as measures of prediction accuracy for the undertaken study. The results show that an RBFN using fuzzy C means performs better than an RBFN using hard K-means. This study uses data on web applications from the Tukutuku database. © 2010 Praise Worthy Prize S.r.l.