Navegando por Assunto "Machine learning"

Agora exibindo 1 - 2 de 2
  • Imagem de Miniatura
    Item
    Finding reduced Raman spectroscopy fingerprint of skin samples for melanoma diagnosis through machine learning
    (Elsevier) Araújo, Daniella Castro; Veloso, Adriano Alonso; Oliveira Filho, Renato Santos de; Giraud, Marie-Noelle; Raniero, Leandro José; Ferreira, Lydia Masako; Bitar, Renata Andrade
    Early-stage detection of cutaneous melanoma can vastly increase the chances of cure. Excision biopsy followed by histological examination is considered the gold standard for diagnosing the disease, but requires long high- cost processing time, and may be biased, as it involves qualitative assessment by a professional. In this paper, we present a new machine learning approach using raw data for skin Raman spectra as input. The approach is highly efficient for classifying benign versus malignant skin lesions (AUC 0.98, 95% CI 0.97–0.99). Furthermore, we present a high-performance model (AUC 0.97, 95% CI 0.95–0.98) using a miniaturized spectral range (896–1039 cm− 1), thus demonstrating that only a single fragment of the biological fingerprint Raman region is needed for producing an accurate diagnosis. These findings could favor the future development of a cheaper and dedicated Raman spectrometer for fast and accurate cancer diagnosis.
  • Imagem de Miniatura
    Item
    Saliva FTIR Spectra and Machine Learning for Autism Spectrum Disorder Diagnosis-Preliminary Study
    (IEEE) Pinto, Mayara Moniz Vieira; Arisawa, Emilia Angela Lo Schiavo; Raniero, Leandro José; Bhattacharjee, Tanmoy
    The diagnosis of Autism Spectrum Disorder (ASD) remains a challenge due to the lack of specific tests and biological markers. ASD is a neurodevelopmental disorder that affects in- dividuals throughout their lives, and its diagnosis allows access to treatments that improve their prognosis. Saliva analysis by Fourier Transform Infrared Spectroscopy (FTIR), which was not previ- ously reported, appears to be a promising diagnostic tool for ASD. This study acquired spectra from samples of 19 ASD and 19 control children. Spectral signatures suggest the dominance of protein secondary structures, β-pleated sheet and α-helix structures in ASD and control children, respectively. Support Vector Machine (SVM) gave the best diagnosis, with sensitivity, precision, and specificity being 92%, 94%, and 95%, respectively. Shapley values analysis to understand the impact of spectral features on the SVM classifier identified β-pleated and β-turn sheets as responsible for classification. Results indicate the potential of saliva-based FTIR for autism diagnosis, warranting a large-scale trial.