A Deep Learning Method for Foot Progression Angle Detection in Plantar Pressure Images

A method for cabbage root posture recognition based on YOLOv5s

Efficient, non-destructive cabbage harvesting is crucial for preserving its flavor and quality. Current cabbage harvesting mainly relies on mechanized automatic picking methods. However, a notable deficiency in most existing cabbage harvesting devices is the absence of a root posture recognition system to promptly adjust the root posture, consequently impacting the accuracy of root cutting during harvesting. To address this issue, this study introduces a cabbage root posture recognition method that combines deep learning with traditional image processing algorithms. Preliminary detection of the main root Region of Interest (ROI) areas of the cabbage is carried out through the YOLOv5s deep learning model. Subsequently, traditional image processing methods, the Graham algorithm, and the method of calculating the minimum circumscribed rectangle are employed to specifically detect the inclination angle of cabbage roots. This approach effectively addresses the difficulty in calculating the inclination angle of roots caused by occlusion from outer leaves. The results demonstrate that the precision and recall of this method are 98.7 % and 98.6 %, respectively, with an average absolute error of 0.80° and an average relative error of 1.34 % in posture. Using this method as a reference for mechanical harvesting can effectively mitigate cabbage damage rates.

Testing protocols and measurement techniques when using pressure sensors for sport and health applications: A comparative review

Plantar pressure measurement systems are routinely used in sports and health applications to assess locomotion. The purpose of this review is to describe and critically discuss: (a) applications of the pressure measurement systems in sport and healthcare, (b) testing protocols and considerations for clinical gait analysis, (c) clinical recommendations for interpreting plantar pressure data, (d) calibration procedures and their accuracy, and (e) the future of pressure sensor data analysis. Rigid pressure platforms are typically used to measure plantar pressures for the assessment of foot function during standing and walking, particularly when barefoot, and are the most accurate for measuring plantar pressures. For reliable data, two step protocol prior to contacting the pressure plate is recommended. In-shoe systems are most suitable for measuring plantar pressures in the field during daily living or dynamic sporting movements as they are often wireless and can measure multiple steps. They are the most suitable equipment to assess the effects of footwear and orthotics on plantar pressures. However, they typically have lower spatial resolution and sampling frequency than platform systems. Users of pressure measurement systems need to consider the suitability of the calibration procedures for their chosen application when selecting and using a pressure measurement system. For some applications, a bespoke calibration procedure is required to improve validity and reliability of the pressure measurement system. The testing machines that are commonly used for dynamic calibration of pressure measurement systems frequently have loading rates of less than even those found in walking, so the development of testing protocols that truly measure the loading rates found in many sporting movements are required. There is clear potential for AI techniques to assist in the analysis and interpretation of plantar pressure data to enable the more complete use of pressure system data in clinical diagnoses and monitoring.

A systematic review of the application of deep learning techniques in the physiotherapeutic therapy of musculoskeletal pathologies

Physiotherapy is a critical area of healthcare that involves the assessment and treatment of physical disabilities and injuries. The use of Artificial Intelligence (AI) in physiotherapy has gained significant attention due to its potential to enhance the accuracy and effectiveness of clinical decision-making and treatment outcomes. Nevertheless, it is still a rather innovative field of application of these techniques and there is a need to find what aspects are highly developed and what possible job niches can be exploited. This systematic review aims to evaluate the current state of research on the use of a particular AI called deep learning models in physiotherapy and identify the key trends, challenges, and opportunities in this field. The findings of this review, conducted following the PRISMA guidelines, provide valuable insights for researchers and clinicians. The most relevant databases included were PubMed, Web of Science, Scopus, Astrophysics Data System, and Central Citation Export. Inclusion and exclusion criteria were established to determine which items would be considered for further review. These criteria were used to screen the items during the first and second peer review processes. A set of quality criteria was developed to select the papers obtained after the second screening. Finally, of the 214 initial papers, 23 studies were selected. From our analysis of the selected articles, we can draw the following conclusions: Concerning deep learning models, innovation is primarily seen in the adoption of hybrid models, with convolutional models being extensively utilized. In terms of data, it's unsurprising that body signals and images are predominantly used. However, texts and structured data present promising avenues for groundbreaking work in the field. Additionally, medical tests that involve the collection of 3D images, Functional Movement Screening, or thermographies emerge as novel areas to explore applications within the scope of physiotherapy.

Artificial intelligence in foot and ankle surgery: current concepts

The twenty-first century has proven that data are the new gold. Artificial intelligence (AI) driven technologies might potentially change the clinical practice in all medical specialities, including orthopedic surgery. AI has a broad spectrum of subcomponents, including machine learning, which consists of a subdivision called deep learning. AI has the potential to increase healthcare delivery, improve indications and interventions, and minimize errors. In orthopedic surgery. AI supports the surgeon in the evaluation of radiological images, training of surgical residents, and excellent performance of machine-assisted surgery. The AI algorithms improve the administrative and management processes of hospitals and clinics, electronic healthcare databases, monitoring the outcomes, and safety controls. AI models are being developed in nearly all orthopedic subspecialties, including arthroscopy, arthroplasty, tumor, spinal and pediatric surgery. The present study discusses current applications, limitations, and future prospective of AI in foot and ankle surgery.

Commercially available pressure sensors for sport and health applications: A comparative review

Pressure measurement systems have numerous applications in healthcare and sport. The purpose of this review is to: (a) describe the brief history of the development of pressure sensors for clinical and sport applications, (b) discuss the design requirements for pressure measurement systems for different applications, (c) critique the suitability, reliability, and validity of commercial pressure measurement systems, and (d) suggest future directions for the development of pressure measurements systems in this area. Commercial pressure measurement systems generally use capacitive or resistive sensors, and typically capacitive sensors have been reported to be more valid and reliable than resistive sensors for prolonged use. It is important to acknowledge, however, that the selection of sensors is contingent upon the specific application requirements. Recent improvements in sensor and wireless technology and computational power have resulted in systems that have higher sensor density and sampling frequency with improved usability - thinner, lighter platforms, some of which are wireless, and reduced the obtrusiveness of in-shoe systems due to wireless data transmission and smaller data-logger and control units. Future developments of pressure sensors should focus on the design of systems that can measure or accurately predict shear stresses in conjunction with pressure, as it is thought the combination of both contributes to the development of pressure ulcers and diabetic plantar ulcers. The focus for the development of in-shoe pressure measurement systems is to minimise any potential interference to the patient or athlete, and to reduce power consumption of the wireless systems to improve the battery life, so these systems can be used to monitor daily activity. A potential solution to reduce the obtrusiveness of in-shoe systems include thin flexible pressure sensors which can be incorporated into socks. Although some experimental systems are available further work is needed to improve their validity and reliability.

Advancements in Artificial Intelligence for Foot and Ankle Surgery: A Systematic Review

Background

There has been a rapid increase in research applying artificial intelligence (AI) to various subspecialties of orthopaedic surgery, including foot and ankle surgery. The purpose of this systematic review is to (1) characterize the topics and objectives of studies using AI in foot and ankle surgery, (2) evaluate the performance of their models, and (3) evaluate their validity (internal or external validation).

Methods

A systematic literature review was conducted using PubMed/MEDLINE and Embase databases in December 2022. All studies that used AI or its subsets machine learning (ML) and deep learning (DL) in the setting of foot and ankle surgery relevant to orthopaedic surgeons were included. Studies were evaluated for their demographics, subject area, outcomes of interest, model(s) tested, model(s)' performance, and validity (internal or external).

Results

A total of 31 studies met inclusion criteria: 14 studies investigated AI for image interpretation, 13 studies investigated AI for clinical predictions, and 4 studies were grouped as "other." Studies commonly explored AI for ankle fractures, calcaneus fractures, hallux valgus, Achilles tendon pathologies, plantar fasciitis, and sports injuries. For studies reporting the area under the receiver operating characteristic curve (AUC), AUCs ranged from 0.64 (poor) to 0.99 (excellent). Two studies (6.45%) reported external validation.

Conclusion

Applications of AI in the field of foot and ankle surgery are expanding, particularly for image interpretation and clinical predictions. Current model performances range from poor to excellent, and most studies lack external validation, demonstrating a need for further research prior to deploying AI-based clinical applications.

Level of Evidence

Level III, retrospective cohort study.

Decision Tree-Based Foot Orthosis Prescription for Patients with Pes Planus

Pes planus, one of the most common foot deformities, includes the loss of the medial arch, misalignment of the rearfoot, and abduction of the forefoot, which negatively affects posture and gait. Foot orthosis, which is effective in normalizing the arch and providing stability during walking, is prescribed for the purpose of treatment and correction. Currently, machine learning technology for classifying and diagnosing foot types is being developed, but it has not yet been applied to the prescription of foot orthosis for the treatment and management of pes planus. Thus, the aim of this study is to propose a model that can prescribe a customized foot orthosis to patients with pes planus by learning from and analyzing various clinical data based on a decision tree algorithm called classification and regressing tree (CART). A total of 8 parameters were selected based on the feature importance, and 15 rules for the prescription of foot orthosis were generated. The proposed model based on the CART algorithm achieved an accuracy of 80.16%. This result suggests that the CART model developed in this study can provide adequate help to clinicians in prescribing foot orthosis easily and accurately for patients with pes planus. In the future, we plan to acquire more clinical data and develop a model that can prescribe more accurate and stable foot orthosis using various machine learning technologies.

Deep Learning-Based System for Preoperative Safety Management in Cataract Surgery

An artificial intelligence-based system was implemented for preoperative safety management in cataract surgery, including facial recognition, laterality (right and left eye) confirmation, and intraocular lens (IOL) parameter verification. A deep-learning model was constructed with a face identification development kit for facial recognition, the You Only Look Once Version 3 (YOLOv3) algorithm for laterality confirmation, and the Visual Geometry Group-16 (VGG-16) for IOL parameter verification. In 171 patients who were undergoing phacoemulsification and IOL implantation, a mobile device (iPad mini, Apple Inc.) camera was used to capture patients’ faces, location of surgical drape aperture, and IOL parameter descriptions on the packages, which were then checked with the information stored in the referral database. The authentication rates on the first attempt and after repeated attempts were 92.0% and 96.3% for facial recognition, 82.5% and 98.2% for laterality confirmation, and 67.4% and 88.9% for IOL parameter verification, respectively. After authentication, both the false rejection rate and the false acceptance rate were 0% for all three parameters. An artificial intelligence-based system for preoperative safety management was implemented in real cataract surgery with a passable authentication rate and very high accuracy.

Machine-learned-based prediction of lower extremity overuse injuries using pressure plates

Although running has many benefits for both the physical and mental health, it also involves the risk of injuries which results in negative physical, psychological and economical consequences. Those injuries are often linked to specific running biomechanical parameters such as the pressure pattern of the foot while running, and they could potentially be indicative for future injuries. Previous studies focus solely on some specific type of running injury and are often only applicable to a gender or running-experience specific population. The purpose of this study is, for both male and female, first-year students, (i) to predict the development of a lower extremity overuse injury in the next 6 months based on foot pressure measurements from a pressure plate and (ii) to identify the predictive loading features. For the first objective, we developed a machine learning pipeline that analyzes foot pressure measurements and predicts whether a lower extremity overuse injury is likely to occur with an AUC of 0.639 and a Brier score of 0.201. For the second objective, we found that the higher pressures exerted on the forefoot are the most predictive for lower extremity overuse injuries and that foot areas from both the lateral and the medial side are needed. Furthermore, there are two kinds of predictive features: the angle of the FFT coefficients and the coefficients of the autoregressive AR process. However, these features are not interpretable in terms of the running biomechanics, limiting its practical use for injury prevention.