Machine learning to mechanically assess 2D and 3D biomimetic electrospun scaffolds for tissue engineering applications: Between the predictability and the interpretability.

Currently, the use of autografts is the gold standard for the replacement of many damaged biological tissues. However, this practice presents disadvantages that can be mitigated through tissue-engineered implants. The aim of this study is to explore how machine learning can mechanically evaluate 2D and 3D polyvinyl alcohol (PVA) electrospun scaffolds (one twisted filament, 3 twisted filament and 3 twisted/braided filament scaffolds) for their use in different tissue engineering applications. Crosslinked and non-crosslinked scaffolds were fabricated and mechanically characterised, in dry/wet conditions and under longitudinal/transverse loading, using tensile testing. 28 machine learning models (ML) were used to predict the mechanical properties of the scaffolds. 4 exogenous variables (structure, environmental condition, crosslinking and direction of the load) were used to predict 2 endogenous variables (Young's modulus and ultimate tensile strength). ML models were able to identify 6 structures and testing conditions with comparable Young's modulus and ultimate tensile strength to ligamentous tissue, skin tissue, oral and nasal tissue, and renal tissue. This novel study proved that Classification and Regression Trees (CART) models were an innovative and easy to interpret tool to identify biomimetic electrospun structures; however, Cubist and Support Vector Machine (SVM) models were the most accurate, with R2 of 0.93 and 0.8, to predict the ultimate tensile strength and Young's modulus, respectively. This approach can be implemented to optimise the manufacturing process in different applications.

Can we achieve biomimetic electrospun scaffolds with gelatin alone?

Introduction: Gelatin is a natural polymer commonly used in biomedical applications in combination with other materials due to its high biocompatibility, biodegradability, and similarity to collagen, principal protein of the extracellular matrix (ECM). The aim of this study was to evaluate the suitability of gelatin as the sole material to manufacture tissue engineering scaffolds by electrospinning. Methods: Gelatin was electrospun in nine different concentrations onto a rotating collector and the resulting scaffold's mechanical properties, morphology and topography were assessed using mechanical testing, scanning electron microscopy and white light interferometry, respectively. After characterizing the scaffolds, the effects of the concentration of the solvents and crosslinking agent were statistically evaluated with multivariate analysis of variance and linear regressions. Results: Fiber diameter and inter-fiber separation increased significantly when the concentration of the solvents, acetic acid (HAc) and dimethyl sulfoxide (DMSO), increased. The roughness of the scaffolds decreased as the concentration of dimethyl sulfoxide increased. The mechanical properties were significantly affected by the DMSO concentration. Immersed crosslinked scaffolds did not degrade until day 28. The manufactured gelatin-based electrospun scaffolds presented comparable mechanical properties to many human tissues such as trabecular bone, gingiva, nasal periosteum, oesophagus and liver tissue. Discussion: This study revealed for the first time that biomimetic electrospun scaffolds with gelatin alone can be produced for a significant number of human tissues by appropriately setting up the levels of factors and their interactions. These findings also extend statistical relationships to a form that would be an excellent starting point for future research that could optimize factors and interactions using both traditional statistics and machine learning techniques to further develop specific human tissue.

Innovative intelligent insole system reduces diabetic foot ulcer recurrence at plantar sites: a prospective, randomised, proof-of-concept study.

BACKGROUND: Prevention of diabetic foot ulcer recurrence in high risk patients, using current standard of care methods, remains a challenge. We hypothesised that an innovative intelligent insole system would be effective in reducing diabetic foot ulcer recurrence in such patients. METHODS: In this prospective, randomised, proof-of-concept study, patients with diabetes, and with peripheral neuropathy and a recent history of plantar foot ulceration were recruited from two multidisciplinary outpatient diabetic foot clinics in the UK, and were randomly assigned to either intervention or control. All patients received an insole system, which measured plantar pressure continuously during daily life. The intervention group received audiovisual alerts via a smartwatch linked to the insole system and offloading instructions when aberrant pressures were detected; the control group did not receive any alerts. The primary outcome was plantar foot ulcer occurrence within 18 months. This trial is registered with ISRCTN, ISRCTN05585501, and is closed to accrual and complete. FINDINGS: Between March 18, 2014, and Dec 20, 2016, 90 patients were recruited and consented to the study, and 58 completed the study. At follow-up, ten ulcers from 8638 person-days were recorded in the control group and four ulcers from 11 835 person-days in the intervention group: a 71% reduction in ulcer incidence in the intervention group compared with the control group (incidence rate ratio 0·29, 95% CI, 0·09-0·93; p=0·037). The number of patients who ulcerated was similar between groups (six of 26 [control group] vs four of 32 [intervention group]; p=0·29); however, individual plantar sites ulcerated more often in the control group (ten of 416) than in the intervention group (four of 512; p=0·047). In an exploratory analysis of good compliers (n=40), ulcer incidence was reduced by 86% in the intervention group versus control group (incidence rate ratio 0·14, 95% CI 0·03-0·63; p=0·011). In the exploratory analysis, plantar callus severity (change from baseline to 6 months) was greater in re-ulcerating patients (6·5, IQR 4·0-8·3) than non-re-ulcerating patients (2·0, 0·0-4·8; p=0·040). INTERPRETATION: To our knowledge, this study is the first to show that continuous plantar pressure monitoring and dynamic offloading guidance, provided by an innovative intelligent insole system, can lead to a reduction in diabetic foot ulcer site recurrence. FUNDING: Diabetes UK and Orpyx Medical Technologies.

Gait-related strategies for the prevention of plantar ulcer development in the high risk foot.

High plantar pressures lead to ulceration in the diabetic foot, particularly in the forefoot region around the metatarsal heads. High plantar pressures persist during gait due to factors such as peripheral neuropathy, foot deformities, limited ankle dorsi flexion range of motion and reduced plantar tissue thickness. Strategies impinging upon gait such as the use of appropriate therapeutic footwear, custom-moulded insoles and injectable silicone can help to reduce plantar pressures and attenuate the risk for ulceration. Shoes adapted with external rocker profiles facilitate plantar flexion and restrict sagittal plane motion of the metatarsophalangeal joint, reducing pressures in the region of the metatarsal heads. Insoles custom-moulded to patient's feet help to reduce plantar pressures and minimise the risk of ulceration in the forefoot region. The loss of subcutaneous fat tissue in the diabetic foot enhances bony prominences and predisposes the foot to high-pressure areas. Silicone is a biocompatible material that can be safely injected into plantar soft tissue to augment tissue thickness and prevent the development of ulceration. This enhancement to the subcutaneous layer is remarkably well retained and is a generally well-adopted procedure in the clinical setting.