Publications by authors named "Thales Levi Azevedo Valente"
Background And Objective: Amblyopia is a public health problem, and strabismus is its primary cause. Our objective is to evaluate the concordance of the diagnosis of strabismus between strabismus expert ophthalmologist and the mhealth application developed for this purpose.
Methods: We evaluated the concordance of the diagnosis of strabismus between the expert ophthalmologist and the mhealth application by screening 224 children and adolescents in the 5-15 years age group, with snapshots of patients' eyes and their analysis thereof.
Spinal cord detection in planning CT for radiotherapy through adaptive template matching, IMSLIC and convolutional neural networks.
Comput Methods Programs Biomed
March 2019
Article Synopsis
- * The proposed methodology uses advanced techniques like adaptive template matching, IMSLIC, and convolutional neural networks to automate the detection of the spinal cord in CT images, reducing human error and making the process more efficient.
- * Testing on 36 CT images showed high accuracy (92.55%), specificity (92.87%), and sensitivity (89.23%) for spinal cord detection, indicating that this computational approach is effective for treatment planning in radiotherapy.
Article Synopsis
- * A deep learning technique combined with the particle swarm optimization (PSO) algorithm was employed to optimize neural network hyperparameters without manual adjustments.
- * Testing on CT scans yielded impressive results with 97.62% accuracy, demonstrating the effectiveness of the PSO algorithm in enhancing sensitivity and reducing false positives in nodule classification.
Article Synopsis
- White matter lesions, common in the elderly, can indicate various brain diseases, making early detection crucial, and MRI is a key tool for this due to its detailed imaging capabilities.
- The proposed computational methodology to detect white matter lesions in MRI involves four steps: image acquisition, preprocessing, segmentation, and classification using SLIC0 clustering and convolutional neural networks.
- The methodology showed impressive results with an accuracy of 98.73% and very low false positives, demonstrating its effectiveness for analyzing brain MRI scans.
Detection of mass regions in mammograms by bilateral analysis adapted to breast density using similarity indexes and convolutional neural networks.
Comput Methods Programs Biomed
March 2018
Article Synopsis
- * It utilizes the DDSM database and includes a two-phase methodology: training (classifying breast tissue and regions) and testing (involves various preprocessing steps and false positive reductions).
- * The results indicate high accuracy rates, with over 95% in classifying breast tissue and strong sensitivity and specificity in detecting mass regions, demonstrating the effectiveness of the proposed method.
Background And Objective: Medical image processing can contribute to the detection and diagnosis of human body anomalies, and it represents an important tool to assist in minimizing the degree of uncertainty of any diagnosis, while providing specialists with an additional source of diagnostic information. Strabismus is an anomaly that affects approximately 4% of the population. Strabismus modifies vision such that the eyes do not properly align, influencing binocular vision and depth perception.
View Article and Find Full Text PDF