Application of machine learning algorithms in thermal images for an automatic classification of lumbar sympathetic blocks. (April 2023)

• Record Type: Journal Article
• Authors: Cañada-Soriano, Mar; Bovaira, Maite; García-Vitoria, Carles; Salvador-Palmer, Rosario; Cibrián Ortiz de Anda, Rosa; Moratal, David; Priego-Quesada, José Ignacio
• Is Part Of: Journal of thermal biology. Volume 113(2023)
• Publication Date: 2023-04
• Subjects: Infrared thermography -- Medicine -- Complex regional pain syndrome -- Sympathetic ganglia; ΔTMax The variation for ipsilateral foot difference in maximum temperature -- ΔTMean The variation for ipsilateral foot difference in mean temperature -- ΔSD The variation for ipsilateral foot difference in standard deviation temperature -- ΔAsymMax The asymmetry variation between ipsilateral and contralateral foot in maximum temperature -- ΔAsymMean The asymmetry variation between ipsilateral and contralateral foot in mean temperature -- ΔAsymSD The asymmetry variation between ipsilateral and contralateral foot in standard deviation temperature -- ANN Artificial neural networks -- AUC Area under the curve -- CRPS Complex regional pain syndrome -- ES Cohen effect size -- FOV Field of view -- GUI Graphical user interface -- IRT Infrared thermography -- IFOV Instantaneous field of view -- KNN K-Nearest neighbours -- LSB Lumbar sympathetic block -- ML Machine learning -- NETD Noise equivalent temperature difference -- RF Random forest -- ROI Region of interest -- SVM Support vector machine; Thermobiology -- Periodicals; Temperature -- Periodicals; Biology -- Periodicals; Thermobiologie -- Périodiques; Thermobiology; Periodicals; 571.46
• DOI: 10.1016/j.jtherbio.2023.103523
• Languages: English
• ISSNs: 0306-4565

Abstract

Abstract: Purpose: There are no previous studies developing machine learning algorithms in the classification of lumbar sympathetic blocks (LSBs) performance using infrared thermography data. The objective was to assess the performance of different machine learning algorithms to classify LSBs carried out in patients diagnosed with lower limbs Complex Regional Pain Syndrome as successful or failed based on the evaluation of thermal predictors. Methods: 66 LSBs previously performed and classified by the medical team were evaluated in 24 patients. 11 regions of interest on each plantar foot were selected within the thermal images acquired in the clinical setting. From every region of interest, different thermal predictors were extracted and analysed in three different moments (minutes 4, 5, and 6) along with the baseline time (just after the injection of a local anaesthetic around the sympathetic ganglia). Among them, the thermal variation of the ipsilateral foot and the thermal asymmetry variation between feet at each minute assessed and the starting time for each region of interest, were fed into 4 different machine learning classifiers: an Artificial Neuronal Network, K-Nearest Neighbours, Random Forest, and a Support Vector Machine. Results: All classifiers presented an accuracy and specificity higher than 70%, sensitivity higher than 67%, and AUC higher than 0.73, and the Artificial Neuronal Network classifier performed the best with a maximum accuracy of 88%, sensitivity of 100%, specificity of 84% and AUC of 0.92, using 3 predictors. Conclusion: These results suggest thermal data retrieved from plantar feet combined with a machine learning-based methodology can be an effective tool to automatically classify LSBs performance. Highlights: All machine learning algorithms had an accuracy and specificity higher than 70%. ArTificial Neuronal Network was the best classifier using 3 predictors. Skin temperature asymmetry variation variables of the central heel had the highest contribution in the models. Thermal data retrieved from plantar feet combined with machine learning can automatically classify LSBs performance.