Paper List

期刊: ArXiv Preprint
发布日期: 2026-02-27
BioinformaticsMedical Informatics

Predictive Analytics for Foot Ulcers Using Time-Series Temperature and Pressure Data

Department of Computer Science, Middlesex University London, London, UK

Md. Tanvir Hasan Turja
Figure
Figure
Figure
Figure
Figure

30秒速读

IN SHORT: This paper addresses the critical need for continuous, real-time monitoring of diabetic foot health by developing an unsupervised anomaly detection framework that identifies early ulcer risk from wearable sensor data, overcoming limitations of sporadic clinical evaluations.

核心创新

  • Methodology First comparative study applying both Isolation Forest and KNN algorithms to multimodal foot sensor data (temperature and pressure) for early DFU risk detection.
  • Methodology Development of a comprehensive feature engineering pipeline extracting 15+ physiological features from raw sensor data, including pressure derivatives, temperature variation rates, and gait cycle metrics.
  • Biology Identification of strong correlation (r=0.48) between mean pressure in sensor region 3 and maximum temperature, providing biomechanical evidence for combined sensor monitoring.

主要结论

  • Isolation Forest demonstrated superior sensitivity for detecting subtle anomalies (micro-pressure changes <100 units) with optimized hyperparameters (100 trees, max_samples=0.6, contamination=0.05), making it ideal for early risk detection.
  • KNN/LOF showed higher sensitivity to extreme deviations (temperature spikes >40°C, pressure peaks in January/June 2024) but with increased false positives, suitable for flagging severe cases requiring immediate intervention.
  • Strong biomechanical correlations were identified between pressure and temperature features (max_pressure_pData_3 and max_temp_tData: r=0.41; mean_pressure_pData_3 and max_temp_tData: r=0.48), validating multimodal sensor integration.
研究空白： Current DFU monitoring relies on sporadic clinical evaluations and subjective patient reports, lacking continuous real-time surveillance. Supervised learning approaches require large labeled datasets of ulceration events that are impractical to collect, delaying detection until after clinical diagnosis.

摘要: Diabetic foot ulcers (DFUs) are a severe complication of diabetes, often resulting in significant morbidity. This paper presents a predictive analytics framework utilizing time-series data captured by wearable foot sensors—specifically NTC thin-film thermocouples for temperature measurement and FlexiForce pressure sensors for plantar load monitoring. Data was collected from healthy subjects walking on an instrumented pathway. Unsupervised machine learning algorithms, Isolation Forest and K-Nearest Neighbors (KNN), were applied to detect anomalies that may indicate early ulcer risk. Through rigorous data preprocessing and targeted feature engineering, physiologic patterns were extracted to identify subtle changes in foot temperature and pressure. Results demonstrate Isolation Forest is sensitive to micro-anomalies, while KNN is effective in flagging extreme deviations, albeit at a higher false-positive rate. Strong correlations between temperature and pressure readings support combined sensor monitoring for improved predictive accuracy. These findings provide a basis for real-time diabetic foot health surveillance, aiming to facilitate earlier intervention and reduce DFU incidence.