Diabetic foot ulcers segmentation challenge report: Benchmark and analysis.

Monitoring the healing progress of diabetic foot ulcers is a challenging process. Accurate segmentation of foot ulcers can help podiatrists to quantitatively measure the size of wound regions to assist prediction of healing status. The main challenge in this field is the lack of publicly available manual delineation, which can be time consuming and laborious. Recently, methods based on deep learning have shown excellent results in automatic segmentation of medical images, however, they require large-scale datasets for training, and there is limited consensus on which methods perform the best. The 2022 Diabetic Foot Ulcers segmentation challenge was held in conjunction with the 2022 International Conference on Medical Image Computing and Computer Assisted Intervention, which sought to address these issues and stimulate progress in this research domain. A training set of 2000 images exhibiting diabetic foot ulcers was released with corresponding segmentation ground truth masks. Of the 72 (approved) requests from 47 countries, 26 teams used this data to develop fully automated systems to predict the true segmentation masks on a test set of 2000 images, with the corresponding ground truth segmentation masks kept private. Predictions from participating teams were scored and ranked according to their average Dice similarity coefficient of the ground truth masks and prediction masks. The winning team achieved a Dice of 0.7287 for diabetic foot ulcer segmentation. This challenge has now entered a live leaderboard stage where it serves as a challenging benchmark for diabetic foot ulcer segmentation.

Offloading and adherence through technological advancements: Modern approaches for better foot care in diabetes.

OBJECTIVE: This manuscript aims to provide a review and synthesis of contemporary advancements in footwear, sensor technology for remote monitoring, and digital health, with a focus on improving offloading and measuring and enhancing adherence to offloading in diabetic foot care. METHODS: A narrative literature review was conducted by sourcing peer-reviewed articles, clinical studies, and technological innovations. This paper includes a review of various strategies, from specifically designed footwear, smart insoles and boots to using digital health interventions, which aim to offload plantar pressure and help prevent and manage wounds more effectively by improving the adherence to such offloading. RESULTS: In-house specially made footwear, sensor technologies remotely measuring pressure and weight-bearing activity, exemplified for example, through applications like smart insoles and SmartBoot, and other digital health technologies, show promise in improving offloading and changing patient behaviour towards improving adherence to offloading and facilitating personalised care. This paper introduces the concept of gamification and emotive visual indicators as novel methods to enhance patient engagement. It further discusses the transformative role of digital health technologies in the modern era. CONCLUSIONS: The integration of technology with footwear and offloading devices offers unparallelled opportunities for improving diabetic foot disease management not only through better offloading but also through improved adherence to offloading. These advancements allow healthcare providers to personalise treatment plans more effectively, thereby promising a major improvement in patient outcomes in diabetic foot ulcer healing and prevention.

Acute Effects of Vibrating Insoles on Dynamic Balance and Gait Quality in Individuals With Diabetic Peripheral Neuropathy: A Randomized Crossover Study.

OBJECTIVE: This study investigated the effects of vibrating insoles on dynamic balance and gait quality during level and stair walking and explored the influence of vibration type and frequency in individuals with diabetic peripheral neuropathy (DPN). RESEARCH DESIGN AND METHODS: Twenty-two men with DPN were assessed for gait quality and postural and dynamic balance during walking and stair negotiation using a motion capture system and force plates across seven vibratory insole conditions (Vcs) versus a control (Ctrl) condition (insole without vibration). Vibration was applied during standing and walking tasks, and 15-min rest-stop periods without vibration were interposed between conditions. Repeated measures test conditions were randomized. The primary outcomes were gait speed and dynamic balance. RESULTS: Gait speed during walking significantly improved in all Vcs compared with Ctrl (P < 0.005), with Vc2, Vc4, and Vc6 identified as the most effective. Gait speed increased (reflecting faster walking) during stair ascent and descent in Vc2 (Ctrl vs. Vc2 for ascent 0.447 ± 0.180 vs. 0.517 ± 0.127 m/s; P = 0.037 and descent 0.394 ± 0.170 vs. 0.487 ± 0.125 m/s; P = 0.016), Vc4 (Ctrl vs. Vc4 for ascent 0.447 ± 0.180 vs. 0.482 ± 0.197 m/s; P = 0.047 and descent 0.394 ± 0.170 vs. 0.438 ± 0.181 m/s; P = 0.017), and Vc6 (Ctrl vs. Vc6 for ascent 0.447 ± 0.180 vs. 0.506 ± 0.179 m/s; P = 0.043 and descent 0.394 ± 0.170 vs. 0.463 ± 0.159 m/s; P = 0.026). Postural balance improved during quiet standing with eyes closed in Vc2, Vc4, Vc6, and Vc7 (P < 0.005). CONCLUSIONS: Vibrating insoles are an effective acute strategy for improving postural balance and gait quality during level walking and stair descent in individuals with DPN. These benefits are particularly evident when the entire plantar foot surface is stimulated.

Psychosocial Care for People With Diabetic Neuropathy: Time for Action.

Psychological factors and psychosocial care for individuals with diabetic neuropathy (DN), a common and burdensome complication of diabetes, are important but overlooked areas. In this article we focus on common clinical manifestations of DN, unremitting neuropathic pain, postural instability, and foot complications, and their psychosocial impact, including depression, anxiety, poor sleep quality, and specific problems such as fear of falling and fear of amputation. We also summarize the evidence regarding the negative impact of psychological factors such as depression on DN, self-care tasks, and future health outcomes. The clinical problem of underdetection and undertreatment of psychological problems is described, together with the value of using brief assessments of these in clinical care. We conclude by discussing trial evidence regarding the effectiveness of current pharmacological and nonpharmacological approaches and also future directions for developing and testing new psychological treatments for DN and its clinical manifestations.

Artificial intelligence for automated detection of diabetic foot ulcers: A real-world proof-of-concept clinical evaluation.

OBJECTIVE: Conduct a multicenter proof-of-concept clinical evaluation to assess the accuracy of an artificial intelligence system on a smartphone for automated detection of diabetic foot ulcers. METHODS: The evaluation was undertaken with patients with diabetes (n = 81) from September 2020 to January 2021. A total of 203 foot photographs were collected using a smartphone, analysed using the artificial intelligence system, and compared against expert clinician judgement, with 162 images showing at least one ulcer, and 41 showing no ulcer. Sensitivity and specificity of the system against clinician decisions was determined and inter- and intra-rater reliability analysed. RESULTS: Predictions/decisions made by the system showed excellent sensitivity (0.9157) and high specificity (0.8857). Merging of intersecting predictions improved specificity to 0.9243. High levels of inter- and intra-rater reliability for clinician agreement on the ability of the artificial intelligence system to detect diabetic foot ulcers was also demonstrated (Kα > 0.8000 for all studies, between and within raters). CONCLUSIONS: We demonstrate highly accurate automated diabetic foot ulcer detection using an artificial intelligence system with a low-end smartphone. This is the first key stage in the creation of a fully automated diabetic foot ulcer detection and monitoring system, with these findings underpinning medical device development.

Diabetes foot complications and standardized mortality rate in type 2 diabetes.

AIM: To quantify the impact of foot complications on mortality outcomes in people with type 2 diabetes (T2D), and how routinely measured factors might modulate that risk. MATERIALS AND METHODS: Data for individuals with T2D for 2010-2020, from the Salford Integrated Care Record (Salford, UK), were extracted for laboratory and clinical data, and deaths. Annual expected deaths were taken from Office of National Statistics mortality data. An index of multiple deprivation (IMD) adjusted the standardized mortality ratio (SMR_IMD). Life years lost per death (LYLD) was estimated from the difference between expected and actual deaths. RESULTS: A total of 11 806 T2D patients were included, with 5583 new diagnoses and 3921 deaths during 2010-2020. The number of expected deaths was 2135; after IMD adjustment, there were 2595 expected deaths. Therefore, excess deaths numbered 1326 (SMR_IMD 1.51). No foot complications were evident in n = 9857. This group had an SMR_IMD of 1.13 and 2.74 LYLD. In total, 2979 patients had any foot complication recorded. In this group, the SMD_IMR was 2.29; of these, 2555 (75%) had only one foot complication. Patients with a foot complication showed little difference in percentage HbA1c more than 58 mmol/mol. In multivariate analysis, for those with a foot complication and an albumin-to-creatinine ratio of more than 3 mg/mmol, the odds ratio (OR) for death was 1.93, and for an estimated glomerular filtration rate of less than 60 mL/min/1.73m2 , the OR for death was 1.92. CONCLUSIONS: Patients with T2D but without a foot complication have an SMR_IMD that is only slightly higher than that of the general population. Those diagnosed with a foot complication have a mortality risk that is double that of those without T2D.

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.

Associations between long-term exercise participation and lower limb joint and whole-bone geometry in young and older adults.

Introduction: Features of lower limb bone geometry are associated with movement kinematics and clinical outcomes including fractures and osteoarthritis. Therefore, it is important to identify their determinants. Lower limb geometry changes dramatically during development, partly due to adaptation to the forces experienced during physical activity. However, the effects of adulthood physical activity on lower limb geometry, and subsequent associations with muscle function are relatively unexplored. Methods: 43 adult males were recruited; 10 young (20-35 years) trained i.e., regional to world-class athletes, 12 young sedentary, 10 older (60-75 years) trained and 11 older sedentary. Skeletal hip and lower limb geometry including acetabular coverage and version angle, total and regional femoral torsion, femoral and tibial lateral and frontal bowing, and frontal plane lower limb alignment were assessed using magnetic resonance imaging. Muscle function was assessed recording peak power and force of jumping and hopping using mechanography. Associations between age, training status and geometry were assessed using multiple linear regression, whilst associations between geometry and muscle function were assessed by linear mixed effects models with adjustment for age and training. Results: Trained individuals had 2° (95% CI:0.6°-3.8°; p = 0.009) higher femoral frontal bowing and older individuals had 2.2° (95% CI:0.8°-3.7°; p = 0.005) greater lateral bowing. An age-by-training interaction indicated 4° (95% CI:1.4°-7.1°; p = 0.005) greater acetabular version angle in younger trained individuals only. Lower limb geometry was not associated with muscle function (p > 0.05). Discussion: The ability to alter skeletal geometry via exercise in adulthood appears limited, especially in epiphyseal regions. Furthermore, lower limb geometry does not appear to be associated with muscle function.

A Signal Processing Method for Assessing Ankle Torque with a Custom-Made Electronic Dynamometer in Participants Affected by Diabetic Peripheral Neuropathy.

Portable, custom-made electronic dynamometry for the foot and ankle is a promising assessment method that enables foot and ankle muscle function to be established in healthy participants and those affected by chronic conditions. Diabetic peripheral neuropathy (DPN) can alter foot and ankle muscle function. This study assessed ankle toque in participants with diabetic peripheral neuropathy and healthy participants, with the aim of developing an algorithm for optimizing the precision of data processing and interpretation of the results and to define a reference frame for ankle torque measurement in both healthy participants and those affected by DPN. This paper discloses the software chain and the signal processing methods used for voltage-torque conversion, filtering, offset detection and the muscle effort type identification, which further allowed for a primary statistical report. The full description of the signal processing methods will make our research reproducible. The applied algorithm for signal processing is proposed as a reference frame for ankle torque assessment when using a custom-made electronic dynamometer. While evaluating multiple measurements, our algorithm permits for a more detailed parametrization of the ankle torque results in healthy participants and those affected by DPN.

Acute Effects of Sedentary Behavior on Ankle Torque Assessed with a Custom-Made Electronic Dynamometer.

Inactivity negatively influences general health, and sedentary behaviour is known to impact the musculoskeletal system. The aim of the study was to assess the impact of time spent in active and sedentary behaviour on foot muscle strength. In this observational study, we compared the acute effects of one day of prolonged sitting and one day of low-to-moderate level of activity on ankle torque in one group of eight healthy participants. Peak ankle torque was measured using a portable custom-made electronic dynamometer. Three consecutive maximal voluntary isometric contractions for bilateral plantar flexor and dorsiflexor muscles were captured at different moments in time. The average peak torque significant statistically decreased at 6 h (p = 0.019) in both static and active behaviours, with a higher average peak torque in the active behaviour (p < 0.001). Age, gender, body mass index and average steps did not have any significant influence on the average value of maximal voluntary isometric contraction. The more time participants maintained either static or active behaviour, the less force was observed during ankle torque testation. The static behaviour represented by the sitting position was associated with a higher reduction in the average peak ankle torque during a maximal voluntary isometric contraction when compared to the active behaviour.

A Custom-Made Electronic Dynamometer for Evaluation of Peak Ankle Torque after COVID-19.

The negative effects of SARS-CoV-2 infection on the musculoskeletal system include symptoms of fatigue and sarcopenia. The aim of this study is to assess the impact of COVID-19 on foot muscle strength and evaluate the reproducibility of peak ankle torque measurements in time by using a custom-made electronic dynamometer. In this observational cohort study, we compare two groups of four participants, one exposed to COVID-19 throughout measurements and one unexposed. Peak ankle torque was measured using a portable custom-made electronic dynamometer. Ankle plantar flexor and dorsiflexor muscle strength was captured for both feet at different ankle angles prior and post COVID-19. Average peak torque demonstrated no significant statistical differences between initial and final moment for both groups (p = 0.945). An increase of 4.8%, p = 0.746 was obtained in the group with COVID-19 and a decrease of 1.3%, p = 0.953 was obtained in the group without COVID-19. Multivariate analysis demonstrated no significant differences between the two groups (p = 0.797). There was a very good test−retest reproducibility between the measurements in initial and final moments (ICC = 0.78, p < 0.001). In conclusion, peak torque variability is similar in both COVID-19 and non-COVID-19 groups and the custom-made electronic dynamometer is a reproducible method for repetitive ankle peak torque measurements.

Site-Specific, Critical Threshold Barefoot Peak Plantar Pressure Associated with Diabetic Foot Ulcer History: A Novel Approach to Determine DFU Risk in the Clinical Setting.

Background and Objectives: Barefoot peak plantar pressures (PPPs) are elevated in diabetes patients with neuropathic foot ulcer (DFU) history; however, there is limited reported evidence for a causative link between high barefoot PPP and DFU risk. We aimed to determine, using a simple mat-based methodology, the site-specific, barefoot PPP critical threshold that will identify a plantar site with a previous DFU. Materials and Methods: In a cross-sectional study, barefoot, site-specific PPPs were measured with normal gait for patients with DFU history (n = 21) and healthy controls (n = 12), using a validated carbon footprint system. For each participant, PPP was recorded at twelve distinct plantar sites (1st-5th toes, 1st-5th metatarsal heads (MTHs), midfoot and heel), per right and left foot, resulting in the analysis of n = 504 distinct plantar sites in the diabetes group, and n = 288 sites in the control group. Receiver operator characteristic curve analysis determined the optimal critical threshold for sites with DFU history. Results: Median PPPs for the groups were: diabetes sites with DFU history (n = 32) = 5.0 (3.25-7.5) kg/cm2, diabetes sites without DFU history (n = 472) = 3.25 (2.0-5.0) kg/cm2, control sites (n = 288) = 2.0 (2.0-3.25) kg/cm2; (p < 0.0001). Diabetes sites with elevated PPP (>6 kg/cm2) were six times more likely to have had DFU than diabetes sites with PPP ≤ 6 kg/cm2 (OR = 6.4 (2.8-14.6, 95% CI), p < 0.0001). PPP > 4.1 kg/cm2 was determined as the optimal critical threshold for identifying DFU at a specific plantar site, with sensitivity/specificity = 100%/79% at midfoot; 80%/65% at 5th metatarsal head; 73%/62% at combined midfoot/metatarsal head areas. Conclusions: We have demonstrated, for the first time, a strong, site-specific relationship between elevated barefoot PPP and previous DFU. We have determined a critical, highly-sensitive, barefoot PPP threshold value of >4.1 kg/cm2, which may be easily used to identify sites of previous DFU occurrence and, therefore, increased risk of re-ulceration. This site-specific approach may have implications for how high PPPs should be investigated in future trials.

Neuromuscular dysfunction and exercise training in people with diabetic peripheral neuropathy: A narrative review.

Diabetic peripheral neuropathy (DPN) is a common condition that is associated with neuromuscular dysfunction and peripheral sensory impairment. These deficits predispose patients to sensory and motor system limitations, foot ulcers and a high risk of falls. Exercise training has been proposed as an effective tool to alleviate neural deficits and improve whole-body function. Here we review the effects of DPN on neuromuscular function, the mechanisms underlying this impairment, and the neural and muscular adaptations to exercise training. Muscle dysfunction is an early hallmark of DPN. Deficits in muscle strength, power, mass and a greater fatigability are particularly severe in the lower extremity muscles. Non-enzymatic glycation of motor proteins, impaired excitation-contraction coupling and loss of motor units have been indicated as the main factors underlying muscular dysfunction. Among the exercise-based solutions, aerobic training improves neural structure and function and ameliorates neuropathic signs and symptoms. Resistance training induces marked improvement of muscle performance and may alleviate neuropathic pain. A combination of aerobic and resistance training (i.e., combined training) restores small sensory nerve damage, reduces symptoms, and improves muscle function. The evidence so far suggests that exercise training is highly beneficial and should be included in the standard care for DPN patients.

The gait profile score characterises walking performance impairments in young stroke survivors.

BACKGROUND: The Gait Profile Score (GPS) provides a composite measure of the quality of joint movement during walking, but the relationship between this measure and metabolic cost, temporal (e.g. walking speed) and spatial (e.g. stride length) parameters in stroke survivors has not been reported. RESEARCH QUESTION: The aims of this study were to compare the GPS (paretic, non-paretic, and overall score) of young stroke survivors to the healthy able-bodied control and determine the relationship between the GPS and metabolic cost, temporal (walking speed, stance time asymmetry) and spatial (stride length, stride width, step length asymmetry) parameters in young stroke survivors to understand whether the quality of walking affects walking performance in stroke survivors. METHODS: Thirty-nine young stroke survivors aged between 18 and 65years and 15 healthy age-matched able-bodied controls were recruited from six hospital sites in Wales, UK. Joint range of motion at the pelvis, hip, knee and ankle, and temporal and spatial parameters were measured during walking on level ground at self-selected speed with calculation of the Gait Variable Score and then the GPS. RESULTS: GPS for the paretic leg (9.40° (8.60-10.21) p < 0.001), non-paretic leg (11.42° (10.20-12.63) p < 0.001) and overall score (11.18° (10.26-12.09) p < 0.001)) for stroke survivors were significantly higher than the control (4.25° (3.40-5.10), 5.92° (5.11 (6.73)). All parameters with the exception of step length symmetry ratio correlated moderate to highly with the GPS for the paretic, non-paretic, and/or overall score (ρ = <-0.732 (p < 0.001)). SIGNIFICANCE: The quality of joint movement during walking measured via the GPS is directly related to the speed and efficiency of walking, temporal (stance time symmetry) and spatial (stride length, stride width) parameters in young stroke survivors.

12-Month changes of muscle strength, body composition and physical activity in adults with dystrophinopathies.

PURPOSE: Muscular dystrophy (MD) is an umbrella term for muscle wasting conditions, for which longitudinal changes in function and body composition are well established in children with Duchenne (DMD), however, changes in adults with DMD and Beckers (BMD), respectively, remain poorly reported. This study aims to assess 12-month changes in lower-limb strength, muscle size, body composition and physical activity in adults with Muscular Dystrophy (MD). METHODS: Adult males with Duchenne MD (DMD; N = 15) and Beckers MD (BMD; N = 12) were assessed at baseline and 12-months for body composition (Body fat and lean body mass (LBM)), Isometric maximal voluntary contraction (Knee-Extension (KEMVC) and Plantar-Flexion (PFMVC)) and physical activity (tri-axial accelerometry). RESULTS: 12-Month change in strength was found as -19% (PFMVC) and -14% (KEMVC) in DMD. 12-Month change in strength in BMD, although non-significant, was explained by physical activity (R2=0.532-0.585). Changes in LBM (DMD) and body fat (BMD) were both masked by non-significant changes in body mass. DISCUSSION: 12-Month changes in adults with DMD appear consistent with paediatric populations. Physical activity appears important for muscle function maintenance. Specific monitoring of body composition, and potential co-morbidities, within adults with MD is highlighted.Implications for rehabilitationQuantitative muscle strength assessment shows progressive muscle weakness in adults with Duchenne Muscular Dystrophy is comparable to paediatric reports (-14 to -19%).Physical activity should be encouraged in adults with Beckers Muscular Dystrophy, anything appears better than nothing.Body composition, rather than body mass, should be monitored closely to identify any increase in body fat.

Patient and Provider Perspective of Smart Wearable Technology in Diabetic Foot Ulcer Prevention: A Systematic Review.

Background and Objectives: Smart wearable devices are effective in diabetic foot ulcer (DFU) prevention. However, factors determining their acceptance are poorly understood. This systematic review aims to examine the literature on patient and provider perspectives of smart wearable devices in DFU prevention. Materials and Methods: PubMed, Scopus, and Web of Science were systematically searched up to October 2021. The selected articles were assessed for methodological quality using the quality assessment tool for studies with diverse designs. Results: A total of five articles were identified and described. The methodological quality of the studies ranged from low to moderate. Two studies employed a quantitative study design and focused on the patient perspective, whereas three studies included a mixed, quantitative/qualitative design and explored patient or provider (podiatrist) perspectives. Four studies focused on an insole system and one included a smart sock device. The quantitative studies demonstrated that devices were comfortable, well designed and useful in preventing DFU. One mixed design study reported that patients did not intend to adopt an insole device in its current design because of malfunctions, a lack of comfort. and alert intrusiveness, despite the general perception that the device was a useful tool for foot risk monitoring. Two mixed design studies found that performance expectancy was a predictor of a podiatrist's behavioural intention to recommend an insole device in clinical practice. Disappointing participant experiences negatively impacted the podiatrists' intention to adopt a smart device. The need for additional refinements of the device was indicated by patients and providers before its use in this population. Conclusions: The current evidence about patient and provider perspectives on smart wearable technology is limited by scarce methodological quality and conflicting results. It is, thus, not possible to draw definitive conclusions regarding acceptability of these devices for the prevention of DFU in people with diabetes.

An intelligent insole system with personalised digital feedback reduces foot pressures during daily life: An 18-month randomised controlled trial.

AIMS: High plantar pressure is a major risk factor in the development of diabetic foot ulcers (DFUs) and recent evidence shows plantar pressure feedback reduces DFU recurrence. This study investigated whether continued use of an intelligent insole system by patients at high-risk of DFUs causes a reduction in plantar pressures. METHODS: Forty-six patients with diabetic peripheral neuropathy and previous DFU were randomised to intervention (IG) or control groups (CG). Patients received an intelligent insole system, consisting of pressure-sensing insoles and digital watch. Patients wore the device during all daily activity for 18-months or until ulceration, and integrated pressure was recorded continuously. The device provided high-pressure feedback to IG only via audio-visual-vibrational alerts. High-pressure parameters at the whole foot, forefoot and rearfoot were compared between groups, with multilevel binary logistic regression analysis. RESULTS: CG experienced more high-pressure bouts over time than IG across all areas of the foot (P < 0.05). Differences between groups became apparent >16 weeks of wearing the device. CONCLUSIONS: Continuous plantar pressure feedback via an intelligent insole system reduces number of bouts of high-pressure in patients at high-risk of DFU. These findings suggest that patients were learning which activities generated high-pressure, and pre-emptively offloading to avoid further alerts.

Deep learning in diabetic foot ulcers detection: A comprehensive evaluation.

There has been a substantial amount of research involving computer methods and technology for the detection and recognition of diabetic foot ulcers (DFUs), but there is a lack of systematic comparisons of state-of-the-art deep learning object detection frameworks applied to this problem. DFUC2020 provided participants with a comprehensive dataset consisting of 2,000 images for training and 2,000 images for testing. This paper summarizes the results of DFUC2020 by comparing the deep learning-based algorithms proposed by the winning teams: Faster R-CNN, three variants of Faster R-CNN and an ensemble method; YOLOv3; YOLOv5; EfficientDet; and a new Cascade Attention Network. For each deep learning method, we provide a detailed description of model architecture, parameter settings for training and additional stages including pre-processing, data augmentation and post-processing. We provide a comprehensive evaluation for each method. All the methods required a data augmentation stage to increase the number of images available for training and a post-processing stage to remove false positives. The best performance was obtained from Deformable Convolution, a variant of Faster R-CNN, with a mean average precision (mAP) of 0.6940 and an F1-Score of 0.7434. Finally, we demonstrate that the ensemble method based on different deep learning methods can enhance the F1-Score but not the mAP.

Sensory-Motor Mechanisms Increasing Falls Risk in Diabetic Peripheral Neuropathy.

Diabetic peripheral neuropathy (DPN) is associated with peripheral sensory and motor nerve damage that affects up to half of diabetes patients and is an independent risk factor for falls. Clinical implications of DPN-related falls include injury, psychological distress and physical activity curtailment. This review describes how the sensory and motor deficits associated with DPN underpin biomechanical alterations to the pattern of walking (gait), which contribute to balance impairments underpinning falls. Changes to gait with diabetes occur even before the onset of measurable DPN, but changes become much more marked with DPN. Gait impairments with diabetes and DPN include alterations to walking speed, step length, step width and joint ranges of motion. These alterations also impact the rotational forces around joints known as joint moments, which are reduced as part of a natural strategy to lower the muscular demands of gait to compensate for lower strength capacities due to diabetes and DPN. Muscle weakness and atrophy are most striking in patients with DPN, but also present in non-neuropathic diabetes patients, affecting not only distal muscles of the foot and ankle, but also proximal thigh muscles. Insensate feet with DPN cause a delayed neuromuscular response immediately following foot-ground contact during gait and this is a major factor contributing to increased falls risk. Pronounced balance impairments measured in the gait laboratory are only seen in DPN patients and not non-neuropathic diabetes patients. Self-perception of unsteadiness matches gait laboratory measures and can distinguish between patients with and without DPN. Diabetic foot ulcers and their associated risk factors including insensate feet with DPN and offloading devices further increase falls risk. Falls prevention strategies based on sensory and motor mechanisms should target those most at risk of falls with DPN, with further research needed to optimise interventions.

Sedentary behaviour is an independent predictor of diabetic foot ulcer development: An 8-year prospective study.

AIMS: To prospectively explore the association between sedentary time (SED-time) and the development of diabetic foot ulcer (DFU) in people with diabetic peripheral neuropathy (DPN). METHODS: 175 DPN individuals who attended the annual evaluation for the SAMBA Study (2012-2019) were included. Main outcome measure was the first diagnosis of DFU. SED-time was measured by the PAS 2.1 questionnaire. Nerve function was evaluated by nerve conduction studies. Vascular function was assessed by Ankle-brachial index (ABI) and pedal pulses. Foot deformity and skin dryness were examined by visual inspection. RESULTS: 62 participants (35.5%) developed a DFU during the study. SED-time was significantly higher in people who developed DFUs (12.8 ± 3.0 vs 9.4 ± 3.1 h/day). Logistic regression showed that among several nervous (motor amplitude, OR 0.33, 95% CI, 0.18-0.60; sensory amplitude, 0.85, 0.77-0.94) and vascular parameters (ABI, 0.23, 0.1-0.61; pedal pulses, 2.81, 0.12-0.63) and foot characteristics (deformity, 2.63, 1.30-5.32; skin dryness, 2.04, 0.95-4.37), SED-time was one of the strongest variables contributing to the development of DFUs (2.95, 1.45-6.44). CONCLUSIONS: SED-time is an independent predictor of the risk of DFU in people with DPN. The monitoring of SED-time with strategies aimed at reducing it should be included in the standard care of diabetic patients.