28 December 2016 in Signal, Image and Video Processing
1 September 2026 in Revista CienciaUANL
LMLuis Filiberto Regino Medina
Luis Filiberto Regino Medina
JCJuan Manuel González Calleros
Juan Manuel González Calleros
El cerebro envía señales eléctricas que se miden con electroencefalogramas (EEG) para diagnosticar enfermedades como la epilepsia. Las ondas cerebrales reflejan distintos estados y sus anomalías pueden indicar problemas. La inteligencia artificial (IA) mejora el análisis de EEG, detectando eventos epilépticos con alta precisión, pero debe apoyar al juicio clínico, no reemplazarlo. La colaboración y formación continua son clave para integrar eficazmente la IA en la medicina. Abstract The brain generates electrical signals that are measured with electroencephalograms (EEG) to diagnose disorders such as epilepsy. Brain waves reflect different states, and abnormalities in them may indicate problems. Artificial intelligence (AI) enhances EEG analysis by detecting epileptic events with high accuracy, but it should support clinical judgment, not replace it. Collaboration and continuous training are key to effectively integrating AI into medicine.
1 April 2026 in Biomedical Physics & Engineering Express
In this study, we present a quantitative characterization of skin pigmentation through the analysis of 1005 diffuse reflectance spectra associated with Fitzpatrick skin phototypes (I–VI) of Hispanic individuals. We employed complementary metrics, including the individual typology angle (ITA), the melanosome volume fraction ( ), the structural similarity index measure (SSIM), and the percentage of spectral overlap. Our findings revealed that the average ITA values decreased progressively from in phototype I to in phototype VI, confirming the inverse correlation between the typological angle and the degree of skin pigmentation. The volumetric fraction of melanosomes ( ) – derived from a quadratic relationship with the chromaticity coordinate ( ) and exhibiting a RMSE of 0.0016 – ranged from in phototype I to phototype VI. These results provide a quantitative estimate of intra-phototype variability, which becomes more pronounced in darker skin tones. Furthermore, SSIM values (0.42 to 0.94) and spectral overlap (0.00% to 52.05%) confirmed high structural similarity between adjacent groups (especially between phototypes IV and V) and total divergence between extreme phototypes (I vs VI). Notably, intermediate SSIM values (0.74–0.86) were observed between second-adjacent phototypes (e.g. III vs V), suggesting a gradual spectral continuity across the scale. Finally, the total melanin content ( µ g mg −1 ) was determined from , achieving a correlation coefficient of ρ = 0.996, highlighting the chromatic heterogeneity and morphological stability essential for clinical and technological applications in diverse populations.