Sandra Marcos-Recio
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Sandra Marcos-Recio has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Ana-Belén Gil-González
Universidad de Salamanca
3 shared publications - Andrés Cardona-Mendoza
El Bosque University
3 shared publications - Lautaro Rossi Labianca
Universidad de Salamanca
3 shared publications - Andrés Barrero-Bueno
AIR Institute, Avenida Santiago Madrigal, 39, 37003 Salamanca, Spain
2 shared publications - Sandra Janneth Perdomo Lara
El Bosque University
2 shared publications
Latest Publications
- 29 July 2026
- Published by Ediciones Universidad de Salamanca
- p. 35-39
CYTOLENS: MARCO PRELIMINAR PARA EL APOYO AL CRIBADO Y LA PRIORIZACIÓN DE HALLAZGOS EN CITOLOGÍA CERVICAL DIGITAL MEDIANTE INTELIGENCIA ARTIFICIAL
El proyecto CytoLens, en colaboración entre la Universidad El Bosque y BISITE, propone un marco metodológico para apoyar el cribado citológico del cáncer cervical en patología digital mediante inteligencia artificial a partir de láminas completas digitalizadas (WSI). El trabajo define un pipeline reproducible de control de calidad y generación de parches, y en su estado actual se centra en detectores de instancias tipo YOLO para aportar evidencia localizada y soporte a la revisión experta. Como evolución del marco, se contempla incorporar modelos de aprendizaje por instancias múltiples (MIL) con autoatención para priorización a nivel de lámina y estrategias de pseudoetiquetado con revisión parcial, con evaluación basada en particionado por paciente y criterios de utilidad clínica.
- 29 July 2026
- Published by Ediciones Universidad de Salamanca
- p. 35-39
CYTOLENS: MARCO PRELIMINAR PARA EL APOYO AL CRIBADO Y LA PRIORIZACIÓN DE HALLAZGOS EN CITOLOGÍA CERVICAL DIGITAL MEDIANTE INTELIGENCIA ARTIFICIAL
El proyecto CytoLens, en colaboración entre la Universidad El Bosque y BISITE, propone un marco metodológico para apoyar el cribado citológico del cáncer cervical en patología digital mediante inteligencia artificial a partir de láminas completas digitalizadas (WSI). El trabajo define un pipeline reproducible de control de calidad y generación de parches, y en su estado actual se centra en detectores de instancias tipo YOLO para aportar evidencia localizada y soporte a la revisión experta. Como evolución del marco, se contempla incorporar modelos de aprendizaje por instancias múltiples (MIL) con autoatención para priorización a nivel de lámina y estrategias de pseudoetiquetado con revisión parcial, con evaluación basada en particionado por paciente y criterios de utilidad clínica.
- Open Access
- 2 May 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (9) , 4464
Enhanced Cellular Detection in Cervical Cytopathology: A Systematic Study of YOLO11 Training Paradigms
Automated cellular detection using deep learning is a key strategy for optimising cervical cancer screening by reducing the healthcare workload and inter-observer variability. However, analysing Whole Slide Image (WSI) patches presents challenges such as annotation scarcity, morphological complexity, and class imbalance. This study systematically evaluates YOLO11-n, YOLO11-s, and YOLO11-m to assess the impact of target variable granularity and training paradigms on performance. Four strategies were analysed: independent and multi-class models, each evaluated at both the specific cell label and diagnostic macro-group levels. To ensure clinical robustness, patient-level data partitioning was implemented to prevent data leakage. Performance was measured using precision, recall, and mAP (0.5 and 0.5:0.95). The results reveal critical trade-offs between fine-grained discrimination and model generalisation when varying the architectural complexity and labelling strategies. The findings indicate that diagnostic aggregation improves stability, whereas single-class training optimises specialised detection. These results provide methodological guidelines for designing AI-assisted screening systems and may inform future extensions of WSI-level diagnostic pipelines.
- 2 May 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (9) , 4464
Open AccessEnhanced Cellular Detection in Cervical Cytopathology: A Systematic Study of YOLO11 Training Paradigms
Automated cellular detection using deep learning is a key strategy for optimising cervical cancer screening by reducing the healthcare workload and inter-observer variability. However, analysing Whole Slide Image (WSI) patches presents challenges such as annotation scarcity, morphological complexity, and class imbalance. This study systematically evaluates YOLO11-n, YOLO11-s, and YOLO11-m to assess the impact of target variable granularity and training paradigms on performance. Four strategies were analysed: independent and multi-class models, each evaluated at both the specific cell label and diagnostic macro-group levels. To ensure clinical robustness, patient-level data partitioning was implemented to prevent data leakage. Performance was measured using precision, recall, and mAP (0.5 and 0.5:0.95). The results reveal critical trade-offs between fine-grained discrimination and model generalisation when varying the architectural complexity and labelling strategies. The findings indicate that diagnostic aggregation improves stability, whereas single-class training optimises specialised detection. These results provide methodological guidelines for designing AI-assisted screening systems and may inform future extensions of WSI-level diagnostic pipelines.
- Open Access
- 13 April 2026
- Published by MDPI AG in Preprints.org
Enhanced Cellular Detection in Cervical Cytopathology: A Systematic Study of YOLOv11 Training Paradigms
Automated cellular detection using deep learning is a key strategy for optimising cervical cancer screening by reducing the healthcare workload and inter-observer variability. However, analyzing Whole Slide Image (WSI) patches presents challenges like annotation scarcity, morphological complexity, and class imbalance. This study conducts a systematic evaluation of YOLOv11 (n, s, and m variants) to assess the impact of target variable granularity and training paradigms on performance. Four strategies were analyzed: independent and multi-class models, each evaluated at both specific cell label and diagnostic macro-group levels. To ensure clinical robustness, patient-level data partitioning was implemented to prevent information leakage. Performance was measured using precision, recall, and mAP (0.5 and 0.5:0.95). The results reveal critical trade-offs between fine-grained discrimination and model generalization when varying architectural complexity and labeling strategies. Findings indicate that diagnostic aggregation improves stability, while single-class training optimizes specialized detection. These results provide methodological guidelines for designing AI-assisted screening systems and establish a foundation for integrating YOLOv11 detectors into Multiple Instance Learning (MIL) frameworks at the WSI scale.
- 13 April 2026
- Published by MDPI AG in Preprints.org
Open AccessEnhanced Cellular Detection in Cervical Cytopathology: A Systematic Study of YOLOv11 Training Paradigms
Automated cellular detection using deep learning is a key strategy for optimising cervical cancer screening by reducing the healthcare workload and inter-observer variability. However, analyzing Whole Slide Image (WSI) patches presents challenges like annotation scarcity, morphological complexity, and class imbalance. This study conducts a systematic evaluation of YOLOv11 (n, s, and m variants) to assess the impact of target variable granularity and training paradigms on performance. Four strategies were analyzed: independent and multi-class models, each evaluated at both specific cell label and diagnostic macro-group levels. To ensure clinical robustness, patient-level data partitioning was implemented to prevent information leakage. Performance was measured using precision, recall, and mAP (0.5 and 0.5:0.95). The results reveal critical trade-offs between fine-grained discrimination and model generalization when varying architectural complexity and labeling strategies. Findings indicate that diagnostic aggregation improves stability, while single-class training optimizes specialized detection. These results provide methodological guidelines for designing AI-assisted screening systems and establish a foundation for integrating YOLOv11 detectors into Multiple Instance Learning (MIL) frameworks at the WSI scale.