Deep Learning Technology to Recognize American Sign Language Alphabet

Historically, individuals with hearing impairments have faced neglect, lacking the necessary tools to facilitate effective communication. However, advancements in modern technology have paved the way for the development of various tools and software aimed at improving the quality of life for hearing-disabled individuals. This research paper presents a comprehensive study employing five distinct deep learning models to recognize hand gestures for the American Sign Language (ASL) alphabet. The primary objective of this study was to leverage contemporary technology to bridge the communication gap between hearing-impaired individuals and individuals with no hearing impairment. The models utilized in this research include AlexNet, ConvNeXt, EfficientNet, ResNet-50, and VisionTransformer were trained and tested using an extensive dataset comprising over 87,000 images of the ASL alphabet hand gestures. Numerous experiments were conducted, involving modifications to the architectural design parameters of the models to obtain maximum recognition accuracy. The experimental results of our study revealed that ResNet-50 achieved an exceptional accuracy rate of 99.98%, the highest among all models. EfficientNet attained an accuracy rate of 99.95%, ConvNeXt achieved 99.51% accuracy, AlexNet attained 99.50% accuracy, while VisionTransformer yielded the lowest accuracy of 88.59%.

A Survey of Image-Based Fault Monitoring in Additive Manufacturing: Recent Developments and Future Directions

Additive manufacturing (AM) has emerged as a transformative technology for various industries, enabling the production of complex and customized parts. However, ensuring the quality and reliability of AM parts remains a critical challenge. Thus, image-based fault monitoring has gained significant attention as an efficient approach for detecting and classifying faults in AM processes. This paper presents a comprehensive survey of image-based fault monitoring in AM, focusing on recent developments and future directions. Specifically, the proponents garnered relevant papers from 2019 to 2023, gathering a total of 53 papers. This paper discusses the essential techniques, methodologies, and algorithms employed in image-based fault monitoring. Furthermore, recent developments are explored such as the use of novel image acquisition techniques, algorithms, and methods. In this paper, insights into future directions are provided, such as the need for more robust image processing algorithms, efficient data acquisition and analysis methods, standardized benchmarks and datasets, and more research in fault monitoring. By addressing these challenges and pursuing future directions, image-based fault monitoring in AM can be enhanced, improving quality control, process optimization, and overall manufacturing reliability.

The Definition of the Tibial Sagittal Plane and the Paradox of Imageless Navigation and Robotics: A Cadaveric Study

BACKGROUND

The accuracy, precision, and repeatability by which the tibial sagittal plane can be found with imageless technology is currently unknown. The purpose of this study was to identify any differences between imageless and image-based technology to define the sagittal plane of the tibia.

METHODS

A computed tomography (CT) was obtained of 18 cadavers with the knee fully extended. The surgical trans-epicondylar axis and several tibial rotation references were acquired on the CT scan. After a medial parapatellar approach, the same anatomical landmarks were acquired in vivo. In the horizontal plane, the angle between the surgical trans-epicondylar axis and the tibial rotational axes was assessed.

RESULTS

Highest accuracy was found for posterior cruciate ligament (PCL)-anterior cruciate ligament (ACL, -1.48°, standard deviation [SD] 13.64; imageless), tibial medial condyle (TMC)-tibial lateral condyle (TLC, 1.72°, SD 4.24; image-based), the ACL-medial border of tibial tuberosity (MTT, -2.89°, SD 18.86; image-based). Highest precision was acquired with image-based technology: TMC-TLC (SD 4.24), PCL-ACL (SD 5.86), and PCL-medial third of tibial tuberosity (M3TT, SD 7.10). Excellent intraobserver and interobserver correlation coefficients were observed with image-based technology: PCL-MTT, anterior medial condyle (AMC)-anterior lateral condyle (ALC), and TMC-TLC (Intraobserver and interobserver correlation coefficients 0.90-0.98).

CONCLUSION

The tibial sagittal plane could be defined with highest accuracy, precision, and repeatability on a preoperative CT. Imageless methodology lacked the precision and repeatability of image-based technology. With the current pursuit of high accuracy and precision in total knee arthroplasty, the reference frame used to quantify implant position should be highly accurate and precise as well.

LEVEL OF EVIDENCE

IV, Case Series.

Anatomic Image-Based Classification of Lumbar Intervertebral Disc Pathologies

Introduction

Several minimally invasive spine approaches and techniques have been developed in recent years. While the disease processes affecting the spinal motion segment have remained largely the same, the emerging technologies have changed treatment options radically and not necessarily in an organized fashion. The current diagnostic techniques, also evolving, have helped us appreciate the disease's pathoanatomy in minute details. A comprehensive classification method accounting for all anatomical variations in the disc disease, tailored to treatment options, is necessary. Such a classification will allow the surgeon to choose an appropriate surgical option in a consistent fashion. We feel that our classification system will help the spine surgeon make that important decision consistently, with minimal risk of leaving behind a significant lesion or disrupting an otherwise normal structure of the spinal motion segment. Furthermore, we feel such a comprehensive classification will help surgeons and other caregivers to standardize treatment approaches to the various presentations of disc disease, and apply the evolving technology in an organized fashion.

Purpose

To develop a comprehensive, treatment-orientated classification of the lumbar disc disease.

Materials and Methods

The literature was reviewed for the classification of disc disease. The morphology of the disc disease, the topography of the disc lesion, and the symptom-complex produced by the disc lesion are identified and graded. The features so identified and graded are placed in a matrix. The combinations of the anatomical features and symptoms are then computed as shown in the matrix. The MRI database held in the office was studied to determine the most frequent combinations of the disc disease and symptom complex.

Results

A total of 494 combinations were identified, but most have no clinical relevance. The retrospective study of the clinical data and MRI studies of 93 patients (50 male and 43 female) revealed the most affected motion-segment was L5-S1 (male = 19.3%, and female = 23.8%). The most common patho-anatomy is a globally bulging disc (T3L1), representing 37.6% of the total. The second most common combination is a degenerated disc with central, intra-annular tear T4L1), representing 20.4% of the total. At 11.8%, globally bulging with severe axial pain and moderate radicular pain represented the most common patho-anatomic/clinical classification (T3L1B4R2). The most frequent top 10 patho-anatomic/clinical classifications represented 15.5% of the total.

Conclusion

In light of the multiple surgical options for excision of the herniated lumbar disc, including the conventional and minimally invasive, and the fact that the imaging technology allows spine surgeons to see in great detail, the disease status of each of the components of the spinal motion segment, it is imperative to develop comprehensive classification systems which take account of the unique characteristics of the disease entity and guide treatment strategies. The classification system presented here is fairly complex, but the software technology will be utilized for the classification system along with the most appropriate treatment approach.

Expanding the Capabilities of Nutrition Research and Health Promotion Through Mobile-Based Applications

Mobile-based applications are popular and prevalently used in the US population. Applications focusing on nutrition offer platforms for quantifying and changing behaviors to improve dietary intake. Such behavior changes can intervene in the relation of diet to promote health and prevent disease. Mobile applications offer a safe and convenient way to collect user data and share it back to users, researchers, and to health care providers. Other lifestyle factors like activity, sleep, and sedentary behavior, can also be quantified and included in investigations of how lifestyle is related to health. Yet, challenges in the assessment offered through mobile applications and effectiveness to change behavior still remain, including rigorous evaluation, demonstration of successful health improvement, and participant engagement. The data mobile applications generate, however, expands opportunities for discovery of the integrated and time-based nature of various daily activities in relation to health. This article is a summary of a symposium at Nutrition 2020 Live Online on the role of mobile applications as a tool for nutrition research and health promotion. The types and capabilities of mobile applications, challenges in their evaluation and use in research, and opportunities for the data they generate along with a specific example, are reviewed.

Walking speed measurement technology: A review

Purpose of review

This article presents an overview of the main technologies used to estimate gait parameters, focusing on walking speed (WS).

Recent findings

New wearable and environmental technologies to estimate WS have been developed in the last five years. Wearable technologies refer to sensors attached to parts of the patient's body that capture the kinematics during walking. Alternatively, environmental technologies capture walking patterns using external instrumentation. In this review, wearable and external technologies have been included.From the different works reviewed, external technologies face the challenge of implementation outside controlled facilities; an advantage that wearable technologies have, but have not been fully explored. Additionally, systems that can track WS changes in daily activities, especially at-home assessments, have not been developed.

Summary

Walking speed is a gait parameter that can provide insight into an individual's health status. Image-based, walkways, wearable, and floor-vibrations technologies are the most current used technologies for estimating WS. In this paper, research from the last five years that explore each technology's capabilities on WS estimation and an evaluation of their technical and clinical aspects is presented.

MR electrical properties imaging using a generalized image-based method

PURPOSE

To develop a fast and easy-to-use electrical properties tomography (EPT) method based on a single MR scan, avoiding both the need of a B₁ -map and transceive phase assumption, and that is robust against noise.

THEORY

Derived from Maxwell's equations, conductivity, and permittivity are reconstructed from a new partial differential equation involving the product of the RF fields and its derivatives. This also allows us to clarify and revisit the relevance of common assumptions of MREPT.

METHODS

Our new governing equation is solved using a 3D finite-difference scheme and compared to previous frameworks. The benefits of our method over selected existing MREPT methods are demonstrated for different simulation models, as well as for both an inhomogeneous agar phantom gel and in vivo brain data at 3T.

RESULTS

Simulation and experimental results are illustrated to highlight the merits of the proposed method over existing methods. We show the validity of our algorithm in versatile configurations, with many transition regions notably. Complex admittivity maps are also provided as a complementary MR contrast.

CONCLUSION

Because it avoids time-consuming RF field mapping and generalizes the use of standard MR image for electrical properties reconstruction, this contribution is promising as a new step forward for clinical applications.

Investigation of microvascular morphological measures for skeletal muscle tissue oxygenation by image-based modelling in three dimensions

The supply of oxygen in sufficient quantity is vital for the correct functioning of all organs in the human body, especially for skeletal muscle during exercise. Traditionally, microvascular oxygen supply capability is assessed by the analysis of morphological measures on transverse cross-sections of muscle, e.g. capillary density or capillary-to-fibre ratio. In this work, we investigate the relationship between microvascular structure and muscle tissue oxygenation in mice. Phase contrast imaging was performed using synchrotron radiation computed tomography (SR CT) to visualize red blood cells (RBCs) within the microvasculature in mouse soleus muscle. Image-based mathematical modelling of the oxygen diffusion from the RBCs into the muscle tissue was subsequently performed, as well as a morphometric analysis of the microvasculature. The mean tissue oxygenation was then compared with the morphological measures of the microvasculature. RBC volume fraction and spacing (mean distance of any point in tissue to the closest RBC) emerged as the best predictors for muscle tissue oxygenation, followed by length density (summed RBC length over muscle volume). The two-dimensional measures of capillary density and capillary-to-fibre ratio ranked last. We, therefore, conclude that, in order to assess the states of health of muscle tissue, it is advisable to rely on three-dimensional morphological measures rather than on the traditional two-dimensional measures.

Image-based dynamic quantification and high-accuracy 3D evaluation of canopy structure of plant populations

BACKGROUND AND AIMS

Global agriculture is facing the challenge of a phenotyping bottleneck due to large-scale screening/breeding experiments with improved breeds. Phenotypic analysis with high-throughput, high-accuracy and low-cost technologies has therefore become urgent. Recent advances in image-based 3D reconstruction offer the opportunity of high-throughput phenotyping. The main aim of this study was to quantify and evaluate the canopy structure of plant populations in two and three dimensions based on the multi-view stereo (MVS) approach, and to monitor plant growth and development from seedling stage to fruiting stage.

METHODS

Multi-view images of flat-leaf cucumber, small-leaf pepper and curly-leaf eggplant were obtained by moving a camera around the plant canopy. Three-dimensional point clouds were reconstructed from images based on the MVS approach and were then converted into surfaces with triangular facets. Phenotypic parameters, including leaf length, leaf width, leaf area, plant height and maximum canopy width, were calculated from reconstructed surfaces. Accurate evaluation in 2D and 3D for individual leaves was performed by comparing reconstructed phenotypic parameters with referenced values and by calculating the Hausdorff distance, i.e. the mean distance between two surfaces.

KEY RESULTS

Our analysis demonstrates that there were good agreements in leaf parameters between referenced and estimated values. A high level of overlap was also found between surfaces of image-based reconstructions and laser scanning. Accuracy of 3D reconstruction of curly-leaf plants was relatively lower than that of flat-leaf plants. Plant height of three plants and maximum canopy width of cucumber and pepper showed an increasing trend during the 70 d after transplanting. Maximum canopy width of eggplants reached its peak at the 40th day after transplanting. The larger leaf phenotypic parameters of cucumber were mostly found at the middle-upper leaf position.

CONCLUSIONS

High-accuracy 3D evaluation of reconstruction quality indicated that dynamic capture of the 3D canopy based on the MVS approach can be potentially used in 3D phenotyping for applications in breeding and field management.

A roadmap for the implementation of mHealth innovations for image-based diagnostic support in clinical and public-health settings: a focus on front-line health workers and health-system organizations

BACKGROUND

Diagnostic support for clinicians is a domain of application of mHealth technologies with a slow uptake despite promising opportunities, such as image-based clinical support. The absence of a roadmap for the adoption and implementation of these types of applications is a further obstacle.

OBJECTIVES

This article provides the groundwork for a roadmap to implement image-based support for clinicians, focusing on how to overcome potential barriers affecting front-line users, the health-care organization and the technical system.

METHODS

A consensual approach was used during a two-day roundtable meeting gathering a convenience sample of stakeholders (n = 50) from clinical, research, policymaking and business fields and from different countries. A series of sessions was held including small group discussions followed by reports to the plenary. Session moderators synthesized the reports in a number of theme-specific strategies that were presented to the participants again at the end of the meeting for them to determine their individual priority.

RESULTS

There were four to seven strategies derived from the thematic sessions. Once reviewed and prioritized by the participants some received greater priorities than others. As an example, of the seven strategies related to the front-line users, three received greater priority: the need for any system to significantly add value to the users; the usability of mHealth apps; and the goodness-of-fit into the work flow. Further, three aspects cut across the themes: ease of integration of the mHealth applications; solid ICT infrastructure and support network; and interoperability.

CONCLUSIONS

Research and development in image-based diagnostic pave the way to making health care more accessible and more equitable. The successful implementation of those solutions will necessitate a seamless introduction into routines, adequate technical support and significant added value.

Merging dietary assessment with the adolescent lifestyle

The use of image-based dietary assessment methods shows promise for improving dietary self-report among children. The Technology Assisted Dietary Assessment (TADA) food record application is a self-administered food record specifically designed to address the burden and human error associated with conventional methods of dietary assessment. Users would take images of foods and beverages at all eating occasions using a mobile telephone or mobile device with an integrated camera [e.g. Apple iPhone, Apple iPod Touch (Apple Inc., Cupertino, CA, USA); Nexus One (Google, Mountain View, CA, USA)]. Once the images are taken, the images are transferred to a back-end server for automated analysis. The first step in this process is image analysis (i.e. segmentation, feature extraction and classification), which allows for automated food identification. Portion size estimation is also automated via segmentation and geometric shape template modeling. The results of the automated food identification and volume estimation can be indexed with the Food and Nutrient Database for Dietary Studies to provide a detailed diet analysis for use in epidemiological or intervention studies. Data collected during controlled feeding studies in a camp-like setting have allowed for formative evaluation and validation of the TADA food record application. This review summarises the system design and the evidence-based development of image-based methods for dietary assessment among children.

A bounded iterative closest point method for minimally invasive registration of the femur

This article describes a novel method for image-based, minimally invasive registration of the femur, for application to computer-assisted unicompartmental knee arthroplasty. The method is adapted from the well-known iterative closest point algorithm. By utilising an estimate of the hip centre on both the preoperative model and intraoperative patient anatomy, the proposed 'bounded' iterative closest point algorithm robustly produces accurate varus-valgus and anterior-posterior femoral alignment with minimal distal access requirements. Similar to the original iterative closest point implementation, the bounded iterative closest point algorithm converges monotonically to the closest minimum, and the presented case includes a common method for global minimum identification. The bounded iterative closest point method has shown to have exceptional resistance to noise during feature acquisition through simulations and in vitro plastic bone trials, where its performance is compared to a standard form of the iterative closest point algorithm.