Publications

Peer-reviewed research in occupational health AI, climate intelligence, probabilistic forecasting, and computer vision.

5 publications23 citations
2022|EGU22, European Geosciences Union General Assembly 2022|2 citations

ML-based Probabilistic Prediction of 2m Temperature and Total Precipitation

MA Zaytar, B Zadrozny, C Watson, DS Civitarese, EE Vos, TM Mathonsi, et al.

AI Summary

Presents a probabilistic ML framework for predicting surface temperature and precipitation with uncertainty quantification. Presented at EGU, one of the world's largest geoscience conferences, as part of IBM's Environmental Intelligence Suite research.

Real-World Applications
Weather prediction with uncertaintyFlood and drought early warningSupply chain climate resilienceEnvironmental intelligence platforms
View on Google Scholar
MSc Thesis
2022|MSc Thesis, University of the Witwatersrand, 2022

Learning Level Set Method by Echo State Network for Image Segmentation

TL Mashinini

AI Summary

Proposes a novel approach using Echo State Networks for learning variational level set segmentation as a spatiotemporal method. Compares ESN, RNN, GRU, LSTM, and 3D CNN architectures. Found that leaking rate and spectral radius critically influence ESN performance.

Real-World Applications
Medical image segmentationAutonomous vehicle visionSatellite imagery analysisReal-time object detection
View on WIReDSpace
2021|arXiv preprint arXiv:2102.00085|3 citations

Long-range seasonal forecasting of 2m-temperature with machine learning

EE Vos, A Gritzman, S Makhanya, T Mashinini, CD Watson

AI Summary

Developed ML models for long-range seasonal temperature forecasting, outperforming traditional numerical weather prediction at extended lead times. Published during IBM Research Africa tenure, integrated into climate intelligence workflows.

Real-World Applications
Climate risk assessmentAgricultural planningEnergy demand forecastingInsurance & reinsurance modeling
View on Google Scholar
2019|IFAC-PapersOnLine 52 (14), 117-122|10 citations

Mine workers threshold shift estimation via optimization algorithms for deep recurrent neural networks

MCI Madahana, JED Ekoru, TL Mashinini, OTC Nyandoro

AI Summary

Uses deep recurrent neural networks with optimization algorithms to estimate hearing threshold shifts in mine workers caused by noise exposure. The model predicts permanent hearing damage progression, enabling earlier intervention.

Real-World Applications
Occupational health monitoringMining safety systemsPredictive hearing loss detectionIndustrial noise management
View on Google Scholar
2019|IFAC-PapersOnLine 52 (14), 249-254|8 citations

Noise level policy advising system for mine workers

MCI Madahana, JED Ekoru, TL Mashinini, OTC Nyandoro

AI Summary

Proposes an intelligent policy advising system that recommends noise exposure limits for mine workers. Combines real-time noise monitoring with ML models to generate actionable safety policies that comply with occupational health regulations.

Real-World Applications
Mining regulatory complianceReal-time safety policy generationOccupational noise controlWorkplace health AI systems
View on Google Scholar