ABSTRACT
Crying is an inevitable character trait that occurs throughout the growth of infants, under conditions where the caregiver may have difficulty interpreting the underlying cause of the cry. Crying can be treated as an audio signal that carries a message about the infant's state, such as discomfort, hunger, and sickness. The primary infant caregiver requires traditional ways of understanding these feelings. Failing to understand them correctly can cause severe problems. Several methods attempt to solve this problem; however, proper audio feature representation and classifiers are necessary for better results. This study uses time-, frequency-, and time-frequency-domain feature representations to gain in-depth information from the data. The time-domain features include zero-crossing rate (ZCR) and root mean square (RMS), the frequency-domain feature includes the Mel-spectrogram, and the time-frequency-domain feature includes Mel-frequency cepstral coefficients (MFCCs). Moreover, time-series imaging algorithms are applied to transform 20 MFCC features into images using different algorithms: Gramian angular difference fields, Gramian angular summation fields, Markov transition fields, recurrence plots, and RGB GAF. Then, these features are provided to different machine learning classifiers, such as decision tree, random forest, K nearest neighbors, and bagging. The use of MFCCs, ZCR, and RMS as features achieved high performance, outperforming state of the art (SOTA). Optimal parameters are found via the grid search method using 10-fold cross-validation. Our MFCC-based random forest (RF) classifier approach achieved an accuracy of 96.39%, outperforming SOTA, the scalogram-based shuffleNet classifier, which had an accuracy of 95.17%.
ABSTRACT
To compare the efficacy and safety of leflunomide (LEF)-anti-TNF-alpha combination therapy to methotrexate (MTX)-anti-TNF-alpha combination therapy in a group of patients with active rheumatoid arthritis (RA). We have recruited 120 patients with RA with a high disease activity despite being treated with MTX (15 mg/week) or LEF (20 mg/die) for 3 months, without side effects. In each of these patients, therapy with either MTX or LEF was continued and randomly combined with an anti-TNF-alpha drug: etanercept, infliximab, or adalimumab. Patients were assessed at study entry and at 4, 12, and at 24 weeks. The efficacy endpoints included variations in the DAS28-ESR and the ACR20, ACR50, and ACR70 responses. At each visit, any side-effect was recorded. There were no statistically significant differences in the DAS28 variations and in the ACR responses between the two groups or among the six subgroups. The number of discontinuation due to the appearance of serious side effects was higher, but not statistically significant, in the LEF-anti-TNF-alpha group than in the MTX-anti-TNF-alpha group. Other adverse events that did not necessitate the discontinuation of therapy occurred much more frequently in patients treated with MTX than in those treated with LEF. Anti-TNF-alpha drugs can be used in combination not only with MTX, but also with LEF, with the same probability of achieving significant clinical improvement in RA patients and without a significantly greater risk of serious adverse events. In contrast, it seems that combination therapy with LEF-anti-TNF-alpha is more readily tolerated than combination therapy with MTX-anti-TNF-alpha.
