Telemedicine: emerging e-medicine

Artificial intelligence for telemedicine diabetic retinopathy screening: a review

PURPOSE

This study aims to compare artificial intelligence (AI) systems applied in diabetic retinopathy (DR) teleophthalmology screening, currently deployed systems, fairness initiatives and the challenges for implementation.

METHODS

The review included articles retrieved from PubMed/Medline/EMBASE literature search strategy regarding telemedicine, DR and AI. The screening criteria included human articles in English, Portuguese or Spanish and related to telemedicine and AI for DR screening. The author's affiliations and the study's population income group were classified according to the World Bank Country and Lending Groups.

RESULTS

The literature search yielded a total of 132 articles, and nine were included after full-text assessment. The selected articles were published between 2004 and 2020 and were grouped as telemedicine systems, algorithms, economic analysis and image quality assessment. Four telemedicine systems that perform a quality assessment, image preprocessing and pathological screening were reviewed. A data and post-deployment bias assessment are not performed in any of the algorithms, and none of the studies evaluate the social impact implementations. There is a lack of representativeness in the reviewed articles, with most authors and target populations from high-income countries and no low-income country representation.

CONCLUSIONS

Telemedicine and AI hold great promise for augmenting decision-making in medical care, expanding patient access and enhancing cost-effectiveness. Economic studies and social science analysis are crucial to support the implementation of AI in teleophthalmology screening programs. Promoting fairness and generalizability in automated systems combined with telemedicine screening programs is not straightforward. Improving data representativeness, reducing biases and promoting equity in deployment and post-deployment studies are all critical steps in model development.

From Pixels to Pathology: Employing Computer Vision to Decode Chest Diseases in Medical Images

Radiology has been a pioneer in the healthcare industry's digital transformation, incorporating digital imaging systems like picture archiving and communication system (PACS) and teleradiology over the past thirty years. This shift has reshaped radiology services, positioning the field at a crucial junction for potential evolution into an integrated diagnostic service through artificial intelligence and machine learning. These technologies offer advanced tools for radiology's transformation. The radiology community has advanced computer-aided diagnosis (CAD) tools using machine learning techniques, notably deep learning convolutional neural networks (CNNs), for medical image pattern recognition. However, the integration of CAD tools into clinical practice has been hindered by challenges in workflow integration, unclear business models, and limited clinical benefits, despite development dating back to the 1990s. This comprehensive review focuses on detecting chest-related diseases through techniques like chest X-rays (CXRs), magnetic resonance imaging (MRI), nuclear medicine, and computed tomography (CT) scans. It examines the utilization of computer-aided programs by researchers for disease detection, addressing key areas: the role of computer-aided programs in disease detection advancement, recent developments in MRI, CXR, radioactive tracers, and CT scans for chest disease identification, research gaps for more effective development, and the incorporation of machine learning programs into diagnostic tools.

Autonomic neuropathic symptoms in patients with diabetes: practical tools for screening in daily routine

BACKGROUND

Diabetic autonomic neuropathy (DAN) is a frequent complication in people with diabetes whose screening is often neglected. This study aimed to evaluate DAN through practical tools in people with diabetes in a referral center for diabetes treatment.

METHODS

DAN symptoms and severity were assessed using the Survey of Autonomic Symptoms (SAS) via digital application (app) in patients attended from June 1, 2021, to November 12, 2021. SAS scoring for DAN was performed using established validated cutoffs. The adhesive with cobalt salt color indicator (Neuropad™) was used as a measure of sudomotor dysfunction. Demographical and clinical data were also collected.

RESULTS

Data from 109 participants, 66.9% T2DM, 73.4% female, with a median age of 54.00 (± 20.00) years, were analyzed. Symptomatic DAN was present in 69.7% of participants and was associated with older age (p = 0.002), higher HbA1c (p = 0.043), higher abdominal circumference (p = 0.019), higher BMI (p = 0.013), more likely to have metabolic syndrome (MS) with a 10-fold increased risk, and more frequent association with diabetic peripheral neuropathy (p = 0.005). Sudomotor dysfunction was found in 65 participants with positive Neuropad™ detected in 63.1% of them.

CONCLUSION

The use of SAS through an app proved to be a practical and easy-to-use instrument to document symptoms of DAN in busy clinical practice. The high frequency of symptoms draws attention to the importance of screening this underdiagnosed diabetes complication. The risk factors and comorbidities associated with symptomatic DAN highlight the patients' phenotypes linked to MS that should be targeted for DAN evaluations in larger samples in the community.

Patient experiences receiving rehabilitation care via telehealth: Identifying opportunities for remote care

Use of telehealth has grown substantially in recent times due to the COVID-19 pandemic. Remote care services may greatly benefit patients with disabilities; chronic conditions; and neurological, musculoskeletal, and pain disorders, thereby allowing continuity of rehabilitation care, reducing barriers such as transportation, and minimizing COVID-19 exposure. In March 2020, our rehabilitation hospital, Shirley Ryan AbilityLab, launched a HIPAA-compliant telemedicine program for outpatient and day rehabilitation clinics and telerehabilitation therapy programs. The objective of this study was to examine patients' experiences and satisfaction with telemedicine in the rehabilitation physician practice, including novel virtual multidisciplinary evaluations. The present study examines survey data collected from 157 patients receiving telemedicine services at Shirley Ryan AbilityLab from December 2020-August 2021. Respondents were 61.8% female, predominantly White (82.2%) with ages ranging across the lifespan (69.4% over age 50 years). Diagnostic categories of the respondents included: musculoskeletal conditions 28%, chronic pain 22.3%, localized pain 10.2%, neurological conditions 26.8%, and Parkinson's and movement disorders 12.7%. Survey responses indicate that the telemedicine experiences were positive and well received. The majority of participants found these services easy to use, effective, and safe, and were overall satisfied with the attention and care they received from the providers-even for those who had not previously used telehealth. Respondents identified a variety of benefits, including alleviating financial and travel-related burdens. There were no significant differences in telehealth experiences or satisfaction across the different clinical diagnostic groups. Respondents viewed the integrated physician and rehabilitation therapist telehealth multidisciplinary model favorably, citing positive feedback regarding receiving multiple perspectives and recommendations, feeling like an integrated member of their healthcare team, and having a comprehensive, holistic team approach along with effective communication. These findings support that telemedicine can provide an effective care model in physiatry (physical medicine and rehabilitation) clinics, across different neurological, musculoskeletal, and pain conditions and in multidisciplinary team care settings. The insights provided by the present study expand our understanding of patient experiences with remote care frameworks for rehabilitation care, while controlling for institutional variation, and ultimately will help provide guidance regarding longer term integration of telemedicine in physiatry and multidisciplinary care models.

SAM-X: sorting algorithm for musculoskeletal x-ray radiography

OBJECTIVE

To develop a two-phased deep learning sorting algorithm for post-X-ray image acquisition in order to facilitate large musculoskeletal image datasets according to their anatomical entity.

METHODS

In total, 42,608 unstructured and pseudonymized radiographs were retrieved from the PACS of a musculoskeletal tumor center. In phase 1, imaging data were sorted into 1000 clusters by a self-supervised model. A human-in-the-loop radiologist assigned weak, semantic labels to all clusters and clusters with the same label were merged. Three hundred thirty-two non-musculoskeletal clusters were discarded. In phase 2, the initial model was modified by "injecting" the identified labels into the self-supervised model to train a classifier. To provide statistical significance, data split and cross-validation were applied. The hold-out test set consisted of 50% external data. To gain insight into the model's predictions, Grad-CAMs were calculated.

RESULTS

The self-supervised clustering resulted in a high normalized mutual information of 0.930. The expert radiologist identified 28 musculoskeletal clusters. The modified model achieved a classification accuracy of 96.2% and 96.6% for validation and hold-out test data for predicting the top class, respectively. When considering the top two predicted class labels, an accuracy of 99.7% and 99.6% was accomplished. Grad-CAMs as well as final cluster results underlined the robustness of the proposed method by showing that it focused on similar image regions a human would have considered for categorizing images.

CONCLUSION

For efficient dataset building, we propose an accurate deep learning sorting algorithm for classifying radiographs according to their anatomical entity in the assessment of musculoskeletal diseases.

KEY POINTS

• Classification of large radiograph datasets according to their anatomical entity. • Paramount importance of structuring vast amounts of retrospective data for modern deep learning applications. • Optimization of the radiological workflow and increase in efficiency as well as decrease of time-consuming tasks for radiologists through deep learning.

Digital Technology-Based Telemedicine for the COVID-19 Pandemic

In the year 2020, the coronavirus disease 2019 (COVID-19) crisis intersected with the development and maturation of several digital technologies including the internet of things (IoT) with next-generation 5G networks, artificial intelligence (AI) that uses deep learning, big data analytics, and blockchain and robotic technology, which has resulted in an unprecedented opportunity for the progress of telemedicine. Digital technology-based telemedicine platform has currently been established in many countries, incorporated into clinical workflow with four modes, including "many to one" mode, "one to many" mode, "consultation" mode, and "practical operation" mode, and has shown to be feasible, effective, and efficient in sharing epidemiological data, enabling direct interactions among healthcare providers or patients across distance, minimizing the risk of disease infection, improving the quality of patient care, and preserving healthcare resources. In this state-of-the-art review, we gain insight into the potential benefits of demonstrating telemedicine in the context of a huge health crisis by summarizing the literature related to the use of digital technologies in telemedicine applications. We also outline several new strategies for supporting the use of telemedicine at scale.

State of telemedicine use in pediatric surgery in the USA—where we stand and what we can gain from the COVID-19 pandemic: a scoping review

Background

Within the last decade, the use of telemedicine within in primary care in the USA has greatly expanded; however, use remains uncommon in surgical specialties. The spread of Coronavirus disease 2019 (COVID-19) prompted healthcare institutions to limit in-person contact, resulting in an increase in telemedicine across all specialties, including pediatric surgery. The aims of this review were to evaluate potential barriers that limited the use of telemedicine in pediatric surgery prior to the COVID-19 period and to define how best to incorporate its use into a pediatric surgical practice going forward.

Methods

A scoping review was performed to identify gaps in the literature pertaining to the use of telemedicine within general pediatric surgery in the USA prior to the outbreak of COVID-19. Next, a focused evaluation of the legislative and organizational policies on telemedicine was performed. Lastly, findings were summarized and recommendations for future research developed in the context of understanding and overcoming barriers that have plagued widespread adoption in the past.

Results

Despite evidence of telemedicine being safe and well received by adult surgical patients, a total of only three studies representing original research on the use of telemedicine within pediatric surgery were identified. Legislative and organizational policies regarding telemedicine have been altered in response to COVID-19, likely resulting in an increase in the use of telemedicine among pediatric surgeons.

Conclusions

Telemedicine offers a safe and effective option for patients seeking an alternative to the in-person clinic appointment. The increased utilization of telemedicine during the COVID-19 pandemic will provide an opportunity to learn how best to leverage the technology to decrease disparities and to overcome previous barriers.

Artificial Intelligence for the Future Radiology Diagnostic Service

Radiology historically has been a leader of digital transformation in healthcare. The introduction of digital imaging systems, picture archiving and communication systems (PACS), and teleradiology transformed radiology services over the past 30 years. Radiology is again at the crossroad for the next generation of transformation, possibly evolving as a one-stop integrated diagnostic service. Artificial intelligence and machine learning promise to offer radiology new powerful new digital tools to facilitate the next transformation. The radiology community has been developing computer-aided diagnosis (CAD) tools based on machine learning (ML) over the past 20 years. Among various AI techniques, deep-learning convolutional neural networks (CNN) and its variants have been widely used in medical image pattern recognition. Since the 1990s, many CAD tools and products have been developed. However, clinical adoption has been slow due to a lack of substantial clinical advantages, difficulties integrating into existing workflow, and uncertain business models. This paper proposes three pathways for AI's role in radiology beyond current CNN based capabilities 1) improve the performance of CAD, 2) improve the productivity of radiology service by AI-assisted workflow, and 3) develop radiomics that integrate the data from radiology, pathology, and genomics to facilitate the emergence of a new integrated diagnostic service.

Examining user perceptions of SwallowIT: A pilot study of a new telepractice application for delivering intensive swallowing therapy to head and neck cancer patients

Consumer feedback and end-user perceptions provide important information regarding the clinical acceptability of new telepractice systems. This pilot investigation aimed to evaluate end-user perceptions of a new asynchronous telepractice application, ‘ SwallowIT’, designed to support patients to remotely complete intensive swallowing therapy during curative chemoradiotherapy (CRT) treatment for head and neck cancer (HNC). Insights were sought from 15 patients with oropharyngeal cancer who used SwallowIT to complete supported home swallowing therapy. Perceptions were evaluated via structured questionnaires, completed following initial orientation to SwallowIT and on completion of CRT. Semi-structured phone interviews were conducted ≥3 months post-treatment. The majority of patients reported positive initial perceptions towards SwallowIT for comfort (87%), confidence (87%), motivation (73%) and support (87%). No statistically significant change in perceptions was observed from baseline to end of CRT ( p > 0.05). Thematic analysis of interviews revealed four main themes: the ease of use of SwallowIT, motivating factors, circumstances which made therapy difficult, and personal preferences for service-delivery models. These preliminary findings demonstrate that SwallowIT was well-perceived by the current group of HNC consumers and suggest that SwallowIT may be well-accepted as an alternate service-delivery model for delivering intensive swallowing therapy during CRT.

Expanding the walls of the health care encounter: support and outcomes for patients online

This paper assesses the relationship between patient-health care provider (HCP) interaction and health behaviors. In total, 109 Native American patients diagnosed with diabetes mellitus were enrolled in a Web-based diabetes monitoring system. The system tracks patient-HCP interaction, and in total 924 personal messages were exchanged. These 924 messages contained 6,411 message units that were content analyzed using a nine-category scheme. Patient blood glucose monitoring was found to be related to the frequency of phatic communication, informational social support, and tangible social support messages, as well as messages containing references to personal contact. Finally, person-centered messages proved to be the single best predictor of patient involvement with the telemedicine system (as measured by the number of times the patient logged into the system).

Reconstruction in Warfare Injuries

Traumatic injuries, especially in the combat setting, stress the surgical team that may be sited in a remote forward area, battling against paucity of time, resources and infrastructure. The lone surgeon may be faced with the arduous challenge of saving life. There is seldom thought given to reconstruction in this high-pressure situation. If the patient survives, morbidity for want of reconstruction can be severe and quality of life can suffer significantly. Reconstruction after 3 to 5 days is fraught with complications and usually does compromise outcome in the post-operative phase. The reconstructive surgeon should be involved early in the management as he can provide coverage for large soft tissue defects after aggressive debridement with panache. If the patient is haemodynamically stable, he should be transferred urgently, preferrably by air, to a higher centre with multi-specialty care, especially being equipped with an orthopaedic and trauma reconstructive surgeon. It has been proved beyond doubt that the healing improves significantly and there is marked decrease in morbidity if coverage of wounds is provided early, before colonized wounds get infected.

Remote Hearing Screenings via Telehealth in a Rural Elementary School

Purpose

Telehealth (telepractice) is the provision of health care services using telecommunications. Telehealth technology typically has been employed to increase the level of health care access for consumers living in rural communities. In this way, audiologists can use telehealth to provide services in the rural school systems. This is important because school hearing screening programs are the foundation of educational audiology programs. Therefore, the goal of this study was to determine the feasibility of providing hearing screening services by telehealth technology to school-age children.

Method

Hearing screening services—including otoscopy, pure-tone, and immittance audiometry—were conducted on 32 children in 3rd grade attending an elementary school in rural Utah. Each child received 1 screening on-site and another through telehealth procedures.

Results

Immittance and otoscopy results were identical for on-site and telehealth screening protocols. Five children responded differently to pure-tone stimuli presented by the telehealth protocol than by the on-site protocol. However, no statistically significant difference was found for pure-tone screening results obtained by telehealth or on-site screening procedures (binomial test, p = .37). Likewise, overall screening results obtained by traditional and telehealth procedures were not statistically significant (binomial test, p = .37).

Conclusion

The results of this study suggest that school hearing screenings may be provided using telehealth technology. This study did find that 5 students performed differently to pure-tone screenings administered by the telehealth protocol in contrast to on-site hearing screening services. Further research is necessary to identify factors leading to false responses to pure-tone hearing screening when telehealth technology is used. In addition, telehealth hearing screening protocols should be conducted with participants of different age groups and experiencing a wide range of hearing loss to further clarify the value of telehealth technology.

Telemedicine: its effects on health communication

This article analyzes telemedicine, the use of distant communication technologies within the context of clinical health care, and the effects it has on health communication. The main effect is that telemedicine has the capacity to substantially transform health care in both positive and negative ways and to radically modify personal face-to-face communication (Turner, 2003). This has tremendous implications for health communication scholars in that they can extend the telemedicine debate by integrating fresh insights into more acceptable approaches that will refine and humanize mediated channels of health communication. There are several key areas of telemedicine that need to be discussed (i.e., e-health services, clinical encounters, etc.), all of which are identified in this article. In addition to describing the past and current applications of telemedicine, this article provides a better understanding of unique needs, resources, problems, and opportunities germane to telemedicine services.

A comprehensive portrait of teleradiology in radiology practices: results from the American College of Radiology's 1999 Survey

OBJECTIVE

This article presents a comprehensive portrait of the characteristics of teleradiology systems of radiology practices as of 1999. Our purposes are to help profile a rapidly evolving area of radiology that has been underexamined to date and to provide a baseline with which future findings can be compared.

MATERIALS AND METHODS

In 1999, the American College of Radiology surveyed 970 practices by mail. A response rate of 66% was achieved. Responses were weighted to represent all radiology practices in the United States. Data from nine questions specifically designed to profile the use of teleradiology were analyzed using descriptive statistical methods and multivariate regression analyses.

RESULTS

Seventy-one percent of multiradiologist practices had teleradiology systems in place, using them to interpret 5% of their studies. For solo practices, corresponding statistics were 30% and 14%. Ninety-two percent of multiradiologist practices with teleradiology systems used them for preliminary on-call interpretation. Other major uses included consultation with other radiologists (20%) and primary interpretation of studies (18%). Ninety-five percent of multiradiologist practices with teleradiology systems used them to interpret CT, 84% used them for sonography, 69% for nuclear medicine, 47% for MRI, and 43% for conventional radiographs.

CONCLUSION

Teleradiology had already become a fixture in most practices by 1999, though it was used for only a small fraction of image interpretations. Its widespread presence positioned teleradiology to become a key element of radiology practice nationwide.

Personal health information management system and its application in referral management

We developed a web-based personal health record (PHR) that can be used by patients to collect and manage their health information (e.g., medical history, past surgeries, medications, and allergies), to request self-referrals, and to store a record of their consultations. The PHR also includes a messaging system that can be structured into the workflow of referral management as well as allowing more general communications. A preliminary study was conducted with 61 patients. Thirty-two patients completed a survey in which 85% of respondents were satisfied with the usability and 94% were satisfied with the overall online referral process. The consulting physicians were satisfied with the content of subjects' personal health information and referral problem descriptions and found the information detailed enough to triage all requested referrals. Patients, physicians, and patient care coordinators reported that their communications were enhanced by the system and found the messaging component convenient to use.

Telerehabilitation research: emerging opportunities

The field of clinical rehabilitation is rooted in the premise that carefully planned and delivered therapeutic intervention enhances patient outcomes. Underlying this statement is a deeper scientific reality: The field exists because biosystems (e.g., tissues, cells, organs, persons) are inherently adaptive and can dynamically change as a function of a sequence of inputs (e.g., exercise, pharmaceuticals). The tools of telerehabilitation help minimize the barrier of distance, both of patients to rehabilitative services and of researchers to subject populations. This enhanced access opens up new possibilities for discovering and implementing optimized intervention strategies across the continuum of care. Telecommunications technologies are reviewed from the perspective of systems models of the telerehabilitation process, with a focus on human-technology interface design and a special emphasis on emerging home and mobile technologies. Approaches for providing clinical rehabilitation services through telerehabilitation are addressed, including innovative consumer-centered approaches. Finally, telerehabilitation is proposed as a tool for reinvigorating the rehabilitative bioengineering research enterprise.