Cargando…

The Promise of Semantic Segmentation in Detecting Actinic Keratosis Using Clinical Photography in the Wild

SIMPLE SUMMARY: Understanding the relationship between the skin cancerization field and actinic keratosis (AK) is crucial for identifying high-risk individuals, implementing early interventions, and preventing the progression to more aggressive forms of skin cancer. Currently, the clinical tools for...

Descripción completa

Detalles Bibliográficos
Autores principales: Derekas, Panagiotis, Spyridonos, Panagiota, Likas, Aristidis, Zampeta, Athanasia, Gaitanis, Georgios, Bassukas, Ioannis
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10571759/
https://www.ncbi.nlm.nih.gov/pubmed/37835555
http://dx.doi.org/10.3390/cancers15194861
Descripción
Sumario:SIMPLE SUMMARY: Understanding the relationship between the skin cancerization field and actinic keratosis (AK) is crucial for identifying high-risk individuals, implementing early interventions, and preventing the progression to more aggressive forms of skin cancer. Currently, the clinical tools for grading field cancerization primarily involve assessing AK burden. In addition to their inherent subjectivity, these grading systems are limited by the high degree of AK lesions’ recurrence. The present study proposes a method based on deep learning and semantic segmentation to improve the monitoring of the AK burden in clinical settings with enhanced automation and precision. The experimental results highlight the effectiveness of the proposed method, paving the way for more effective and reliable evaluation, continuous monitoring of condition progression, and assessment of treatment responses. ABSTRACT: AK is a common precancerous skin condition that requires effective detection and treatment monitoring. To improve the monitoring of the AK burden in clinical settings with enhanced automation and precision, the present study evaluates the application of semantic segmentation based on the U-Net architecture (i.e., AKU-Net). AKU-Net employs transfer learning to compensate for the relatively small dataset of annotated images and integrates a recurrent process based on convLSTM to exploit contextual information and address the challenges related to the low contrast and ambiguous boundaries of AK-affected skin regions. We used an annotated dataset of 569 clinical photographs from 115 patients with actinic keratosis to train and evaluate the model. From each photograph, patches of 512 × 512 pixels were extracted using translation lesion boxes that encompassed lesions in different positions and captured different contexts of perilesional skin. In total, 16,488 translation-augmented crops were used for training the model, and 403 lesion center crops were used for testing. To demonstrate the improvements in AK detection, AKU-Net was compared with plain U-Net and U-Net++ architectures. The experimental results highlighted the effectiveness of AKU-Net, improving upon both automation and precision over existing approaches, paving the way for more effective and reliable evaluation of actinic keratosis in clinical settings.