Fast Prediction of Femoral Biomechanics Using Supervised Machine Learning and Statistical Shape Modeling

Finite element (FE) analysis is an important computational tool in biomechanics. However, its adoption into clinical practice has been hampered by its computational complexity and required high technical competences for clinicians. In this paper we propose a supervised learning approach to predict the outcome of the FE analysis. We demonstrate our approach on clinical CT and X-ray femur images for FE predictions ( FEP), with features extracted, respectively, from a statistical shape model and from 2D-based morphometric and density information. Using leave-one-out experiments and sensitivity analysis, comprising a database of 89 clinical cases, our method is capable of predicting the distribution of stress values for a walking loading condition with an average correlation coefficient of 0.984 and 0.976, for CT and X-ray images, respectively. These findings suggest that supervised learning approaches have the potential to leverage the clinical integration of mechanical simulations for the treatment of musculoskeletal conditions.

Real-time adaptive models for the personalized prediction of glycemic profile in type 1 diabetes patients

Prediction of glycemic profile is an important task for both early recognition of hypoglycemia and enhancement of the control algorithms for optimization of insulin infusion rate. Adaptive models for glucose prediction and recognition of hypoglycemia based on statistical and artificial intelligence techniques are presented.

Evaluation of p24-based antiretroviral treatment monitoring in pediatric HIV-1 infection: prediction of the CD4+ T-cell changes between consecutive visits

Worldwide, 700,000 infants are infected annually by HIV-1, most of them in resource-limited settings. Care for these children requires simple, inexpensive tests. We have evaluated HIV-1 p24 antigen for antiretroviral treatment (ART) monitoring in children. p24 by boosted enzyme-linked immunosorbent assay of heated plasma and HIV-1 RNA were measured prospectively in 24 HIV-1-infected children receiving ART. p24 and HIV-1 RNA concentrations and their changes between consecutive visits were related to the respective CD4+ changes. Age at study entry was 7.6 years; follow-up was 47.2 months, yielding 18 visits at an interval of 2.8 months (medians). There were 399 complete visit data sets and 375 interval data sets. Controlling for variation between individuals, there was a positive relationship between concentrations of HIV-1 RNA and p24 (P < 0.0001). While controlling for initial CD4+ count, age, sex, days since start of ART, and days between visits, the relative change in CD4+ count between 2 successive visits was negatively related to the corresponding relative change in HIV-1 RNA (P = 0.009), but not to the initial HIV-1 RNA concentration (P = 0.94). Similarly, we found a negative relationship with the relative change in p24 over the interval (P < 0.0001), whereas the initial p24 concentration showed a trend (P = 0.08). Statistical support for the p24 model and the HIV-1 RNA model was similar. p24 may be an accurate low-cost alternative to monitor ART in pediatric HIV-1 infection.

Prediction of Trabecular Bone Anisotropy from Quantitative Computed Tomography using Supervised Learning and a Novel Morphometric Feature Descriptor

Patient-specific biomechanical models including local bone mineral density and anisotropy have gained importance for assessing musculoskeletal disorders. However the trabecular bone anisotropy captured by high-resolution imaging is only available at the peripheral skeleton in clinical practice. In this work, we propose a supervised learning approach to predict trabecular bone anisotropy that builds on a novel set of pose invariant feature descriptors. The statistical relationship between trabecular bone anisotropy and feature descriptors were learned from a database of pairs of high resolution QCT and clinical QCT reconstructions. On a set of leave-one-out experiments, we compared the accuracy of the proposed approach to previous ones, and report a mean prediction error of 6% for the tensor norm, 6% for the degree of anisotropy and 19◦ for the principal tensor direction. These findings show the potential of the proposed approach to predict trabecular bone anisotropy from clinically available QCT images.