Bello T.

Key Skills

Software

Other

Top Subject Matter

Explainable AI for breast cancer diagnosis (medical imaging)

Machine Learning for fall and bend detection in elderly people

Medical imaging: skin disease image classification

Top Data Types

Image

Text

Document

Top Task Types

Diagnosis

Action Recognition

Classification

Freelancer Overview

Research Experience: Explainable AI for Breast Cancer Diagnosis - University of Ibadan (2024). Brings 6+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Master of Science, University of Ibadan (2024) and Bachelor of Science, Adekunle Ajasin University (2020). AI-training focus includes data types such as Medical and DICOM and labeling workflows including Diagnosis, Action Recognition, and Classification.

Labeling Experience

Research Experience: Skin Disease Image Classification (2025)

OtherClassification

Built a computer vision model for automated classification of skin diseases from dermatological images. Applied data augmentation and transfer learning to improve generalization when working with limited medical imaging datasets. Assessed performance against clinical diagnostic benchmarks to evaluate suitability for resource-constrained healthcare settings.•Performed data preparation and augmentation for skin lesion images.•Implemented transfer learning using CNN architectures for classification.•Validated model predictions using clinical benchmark comparisons.•Prepared results to support potential deployment in limited-resource clinical settings.

2025

Research Experience: Fall & Bend Detection for Elderly People | University of Ibadan (2023–2024)

OtherAction Recognition

Designed and trained a machine learning system to detect fall events and abnormal bending postures in elderly individuals using sensor and image data. Prioritized real-time, low-latency predictions to enable early intervention and reduce fall-related injuries among at-risk populations. Investigated feature engineering and model interpretability to make outputs actionable for caregivers and healthcare providers.•Trained models for fall and bend detection using multimodal sensor/image inputs.•Optimized for low-latency inference suitable for real-time monitoring.•Used interpretability-focused analysis to ensure outputs are understandable and usable.•Evaluated feature sets and model behavior to improve actionable detection quality.

2023

Research Experience: Explainable AI for Breast Cancer Diagnosis - University of Ibadan (2024)

OtherDiagnosis

Developed explainable breast cancer diagnostic models using mammographic image data with EfficientNet and LIME to produce interpretable predictions. Evaluated model accuracy, recall, and AUC metrics while using LIME saliency maps to support clinical transparency and trustworthiness. Focused on generating explanations that align model outputs with medically meaningful diagnostic reasoning for healthcare decision-making.•Built fine-tuned CNN pipelines for breast cancer diagnosis.•Applied LIME to generate interpretable saliency maps for model predictions.•Measured performance using accuracy/recall/AUC and interpretability quality metrics.•Documented and validated results to support clinical decision-making use cases.

2023

Education

Work History