-
Cureus May 2024Non-small cell lung carcinoma (NSCLC) is a prevalent and aggressive form of lung cancer, with a poor prognosis for metastatic disease. Immunotherapy, particularly immune... (Review)
Review
Non-small cell lung carcinoma (NSCLC) is a prevalent and aggressive form of lung cancer, with a poor prognosis for metastatic disease. Immunotherapy, particularly immune checkpoint inhibitors (ICIs), has revolutionized the management of NSCLC, but response rates are highly variable. Identifying reliable predictive biomarkers is crucial to optimize patient selection and treatment outcomes. This systematic review aimed to evaluate the current state of artificial intelligence (AI) and machine learning (ML) applications in predicting the response to immunotherapy in NSCLC. A comprehensive literature search identified 19 studies that met the inclusion criteria. The studies employed diverse AI/ML techniques, including deep learning, artificial neural networks, support vector machines, and gradient boosting methods, applied to various data modalities such as medical imaging, genomic data, clinical variables, and immunohistochemical markers. Several studies demonstrated the ability of AI/ML models to accurately predict immunotherapy response, progression-free survival, and overall survival in NSCLC patients. However, challenges remain in data availability, quality, and interpretability of these models. Efforts have been made to develop interpretable AI/ML techniques, but further research is needed to improve transparency and explainability. Additionally, translating AI/ML models from research settings to clinical practice poses challenges related to regulatory approval, data privacy, and integration into existing healthcare systems. Nonetheless, the successful implementation of AI/ML models could enable personalized treatment strategies, improve treatment outcomes, and reduce unnecessary toxicities and healthcare costs associated with ineffective treatments.
PubMed: 38939246
DOI: 10.7759/cureus.61220 -
Advances in Laboratory Medicine Jun 2024The metaverse is a virtual world that is being developed to allow people to interact with each other and with digital objects in a more immersive way. It involves... (Review)
Review
The metaverse is a virtual world that is being developed to allow people to interact with each other and with digital objects in a more immersive way. It involves the convergence of three major technological trends: telepresence, the digital twin, and blockchain. Telepresence is the ability of people to "be together" in a virtual way while not being close to each other. The digital twin is a virtual, digital equivalent of a patient, a medical device or even a hospital. Blockchain can be used by patients to keep their personal medical records secure. In medicine and healthcare, the metaverse could be used in several ways: (1) virtual medical consultations; (2) medical education and training; (3) patient education; (4) medical research; (5) drug development; (6) therapy and support; (7) laboratory medicine. The metaverse has the potential to enable more personalized, efficient, and accessible healthcare, improving patient outcomes and reducing healthcare costs. However, the implementation of the metaverse in medicine and healthcare will require careful consideration of ethical and privacy concerns, as well as social, technical and regulatory challenges. Overall, the future of the metaverse in healthcare looks bright, but new metaverse-specific laws should be created to help overcome any potential downsides.
PubMed: 38939198
DOI: 10.1515/almed-2023-0124 -
JAMIA Open Jul 2024Absolute risk models estimate an individual's future disease risk over a specified time interval. Applications utilizing server-side risk tooling, the R-based iCARE...
OBJECTIVES
Absolute risk models estimate an individual's future disease risk over a specified time interval. Applications utilizing server-side risk tooling, the R-based iCARE (R-iCARE), to build, validate, and apply absolute risk models, face limitations in portability and privacy due to their need for circulating user data in remote servers for operation. We overcome this by porting iCARE to the web platform.
MATERIALS AND METHODS
We refactored R-iCARE into a Python package (Py-iCARE) and then compiled it to WebAssembly (Wasm-iCARE)-a portable web module, which operates within the privacy of the user's device.
RESULTS
We showcase the portability and privacy of Wasm-iCARE through 2 applications: for researchers to statistically validate risk models and to deliver them to end-users. Both applications run entirely on the client side, requiring no downloads or installations, and keep user data on-device during risk calculation.
CONCLUSIONS
Wasm-iCARE fosters accessible and privacy-preserving risk tools, accelerating their validation and delivery.
PubMed: 38938691
DOI: 10.1093/jamiaopen/ooae055 -
Kidney Medicine Jul 2024In the wake of the coronavirus disease 2019 (COVID-19) pandemic, the United States federal government expanded originating telemedicine sites to include outpatient...
RATIONALE & OBJECTIVE
In the wake of the coronavirus disease 2019 (COVID-19) pandemic, the United States federal government expanded originating telemedicine sites to include outpatient dialysis units. For the first time, nephrology practitioners across the United States could replace face-to-face visits with telemedicine for patients receiving in-center hemodialysis. This study describes patients' perspectives on the use of telemedicine during in-center hemodialysis.
STUDY DESIGN
A qualitative study.
SETTING & PARTICIPANTS
Thirty-two patients from underserved populations (older, less educated, unemployed, persons of color) receiving in-center hemodialysis who used telemedicine with their nephrologist during the COVID-19 pandemic.
ANALYTICAL APPROACH
Telephone semistructured interviews were conducted in English or Spanish. Transcripts were thematically analyzed.
RESULTS
We identified 6 themes with subthemes: adapting to telemedicine (gaining familiarity and confidence, overcoming and resolving technical difficulties, and relying on staff for communication); ensuring availability of the physician (enabling an immediate response to urgent medical needs, providing peace of mind, addressing patient needs adequately, and enhanced attention and contact from physicians); safeguarding against infection (limiting COVID-19 exposures and decreasing use); straining communication and physical interactions (loss of personalized touch, limited physical examination, and unable to reapproach physicians about forgotten issues); maintaining privacy (enhancing privacy and projecting voice enables others to hear); and supporting confidence in telemedicine (requiring established rapport with physicians, clinical stabilty of health, and ability to have in-person visits when necessary).
LIMITATIONS
Interviews were conducted later in the pandemic when some nephrology care providers were using telemedicine infrequently.
CONCLUSIONS
Patients receiving in-center hemodialysis adapted to telemedicine visits by their nephrologists in the context of the COVID-19 pandemic and observed its benefits. However, further considerations regarding communication, privacy, and physical assessments are necessary. Integrating telemedicine into future in-center hemodialysis care using a hybrid approach could potentially build trust, optimize communication, and augment care.
PubMed: 38938646
DOI: 10.1016/j.xkme.2024.100848 -
JACC. Advances Oct 2023Mobile health (mHealth) interventions are increasingly being used for cardiovascular research and physical activity promotion.
BACKGROUND
Mobile health (mHealth) interventions are increasingly being used for cardiovascular research and physical activity promotion.
OBJECTIVES
As a result, the authors aimed to evaluate which features facilitate and impede routine engagement with mobile fitness applications.
METHODS
We distributed a pan-Canadian online questionnaire via the behavioral research platform Prolific.co to evaluate what features are associated with the use and routine engagement (ie, daily or weekly use) of mHealth fitness applications and attitudes about data sharing. Binary logistic regression was used to quantify the association between these endpoints and exploratory factors such as the perceived utility of various mHealth application features.
RESULTS
The survey received 694 responses. Most people were women (62%), the median age was 28 years (range: 18-78 years), and most people reported current use of an mHealth fitness application (48%). The perceived importance of personal health (OR: 2.40; 95% CI: 1.34-4.50) was the factor most associated with the current use of an mHealth fitness application. The feature most associated with routine engagement was the ability to track progress toward a goal (OR: 5.10; 95% CI: 2.73-9.61) while the most significant barrier was the absence of goal customization features (OR: 0.44; 95% CI: 0.25-0.81). The acceptance of sharing health data for research was high (56%), and privacy concerns did not significantly affect routine engagement (OR: 0.81; 95% CI: 0.40-1.77). Results were consistent across race and gender.
CONCLUSIONS
mHealth interventions have the potential to be scaled across populations. Optimizing applications to improve self-monitoring and personalization could increase routine engagement and, thus, user retention and intervention effectiveness.
PubMed: 38938369
DOI: 10.1016/j.jacadv.2023.100613 -
Psychiatric Services (Washington, D.C.) Jun 2024The authors investigated barriers to practices that promote family involvement in mental health services, focusing on individuals with severe mental illness, their... (Review)
Review
OBJECTIVE
The authors investigated barriers to practices that promote family involvement in mental health services, focusing on individuals with severe mental illness, their families, and mental health providers. Additionally, the authors sought to identify strategies to facilitate family involvement in mental health provision to highlight the engagement process in routine practice and propose future directions for organizations to establish a family-friendly environment.
METHODS
Systematic searches for literature published from January 1990 to March 2023 were conducted in PsycInfo, PubMed, CINAHL, Sociological Abstracts, and Scopus databases. Gray literature searches and backward and forward snowballing strategies were also used.
RESULTS
Forty-six articles were reviewed, revealing contextual backgrounds and engagement practices that hindered family involvement. Inconsistencies in family involvement stemmed from organizational culture, societal attitudes, and providers' negating of family expertise. Uncertainty regarding confidentiality policies and the absence of practice guidelines posed challenges for providers. Negative experiences of families within the mental health system along with variable commitment also hampered involvement. Some service users declined family involvement because of privacy concerns and differing expectations regarding the extent of involvement. Promoting a shared culture of family work, integrating practice standards, and engaging in professional development activities emerged as key strategies.
CONCLUSIONS
A gap exists between implementing policies and practices for family involvement in mental health treatment. Without cultural and organizational shifts in support of working with families, the uptake of family involvement practices will remain inadequate. Each stakeholder has different perceptions of the barriers to family involvement, and family involvement will remain elusive without a shared agreement on its importance.
PubMed: 38938096
DOI: 10.1176/appi.ps.20230452 -
Harm Reduction Journal Jun 2024Patients with opioid use disorder (OUD) experience various forms of stigma at the individual, public, and structural levels that can affect how they access and engage...
BACKGROUND
Patients with opioid use disorder (OUD) experience various forms of stigma at the individual, public, and structural levels that can affect how they access and engage with healthcare, particularly with medications for OUD treatment. Telehealth is a relatively new form of care delivery for OUD treatment. As reducing stigma surrounding OUD treatment is critical to address ongoing gaps in care, the aim of this study was to explore how telehealth impacts patient experiences of stigma.
METHODS
In this qualitative study, we interviewed patients with OUD at a single urban academic medical center consisting of multiple primary care and addiction clinics in Oregon, USA. Participants were eligible if they had (1) at least one virtual visit for OUD between March 2020 and December 2021, and (2) a prescription for buprenorphine not exclusively used for chronic pain. We conducted phone interviews between October and December 2022, then recorded, transcribed, dual-coded, and analyzed using reflexive thematic analysis.
RESULTS
The mean age of participants (n = 30) was 40.5 years (range 20-63); 14 were women, 15 were men, and two were transgender, non-binary, or gender-diverse. Participants were 77% white, and 33% had experienced homelessness in the prior six months. We identified four themes regarding how telehealth for OUD treatment shaped patient perceptions of and experiences with stigma at the individual (1), public (2-3), and structural levels (4): (1) Telehealth offers wanted space and improved control over treatment setting; (2) Public stigma and privacy concerns can impact both telehealth and in-person encounters, depending on clinical and personal circumstances; (3) The social distance of telehealth could mitigate or exacerbate perceptions of clinician stigma, depending on both patient and clinician expectations; (4) The increased flexibility of telehealth translated to perceptions of increased clinician trust and respect.
CONCLUSIONS
The forms of stigma experienced by individuals with OUD are complex and multifaceted, as are the ways in which those experiences interact with telehealth-based care. The mixed results of this study support policies allowing for a more individualized, patient-centered approach to care delivery that allows patients a choice over how they receive OUD treatment services.
Topics: Humans; Female; Male; Telemedicine; Adult; Middle Aged; Social Stigma; Opioid-Related Disorders; Young Adult; Qualitative Research; Oregon; Buprenorphine; Opiate Substitution Treatment
PubMed: 38937779
DOI: 10.1186/s12954-024-01043-5 -
Reproductive Health Jun 2024Cervical cancer is the fourth most frequent cancer among women, with 90% of cervical cancer-related deaths occurring in low- and middle-income countries like Cameroon....
BACKGROUND
Cervical cancer is the fourth most frequent cancer among women, with 90% of cervical cancer-related deaths occurring in low- and middle-income countries like Cameroon. Visual inspection with acetic acid is often used in low-resource settings to screen for cervical cancer; however, its accuracy can be limited. To address this issue, the Swiss Federal Institute of Technology Lausanne and the University Hospitals of Geneva are collaborating to develop an automated smartphone-based image classifier that serves as a computer aided diagnosis tool for cancerous lesions. The primary objective of this study is to explore the acceptability and perspectives of women in Dschang regarding the usage of a screening tool for cervical cancer relying on artificial intelligence. A secondary objective is to understand the preferred form and type of information women would like to receive regarding this artificial intelligence-based screening tool.
METHODS
A qualitative methodology was employed to gain better insight into the women's perspectives. Participants, aged between 30 and 49 were invited from both rural and urban regions and semi-structured interviews using a pre-tested interview guide were conducted. The focus groups were divided on the basis of level of education, as well as HPV status. The interviews were audio-recorded, transcribed, and coded using the ATLAS.ti software.
RESULTS
A total of 32 participants took part in the six focus groups, and 38% of participants had a primary level of education. The perspectives identified were classified using an adapted version of the Technology Acceptance Model. Key factors influencing the acceptability of artificial intelligence include privacy concerns, perceived usefulness, and trust in the competence of providers, accuracy of the tool as well as the potential negative impact of smartphones.
CONCLUSION
The results suggest that an artificial intelligence-based screening tool for cervical cancer is mostly acceptable to the women in Dschang. By ensuring patient confidentiality and by providing clear explanations, acceptance can be fostered in the community and uptake of cervical cancer screening can be improved.
TRIAL REGISTRATION
Ethical Cantonal Board of Geneva, Switzerland (CCER, N°2017-0110 and CER-amendment n°4) and Cameroonian National Ethics Committee for Human Health Research (N°2022/12/1518/CE/CNERSH/SP). NCT: 03757299.
Topics: Humans; Female; Uterine Cervical Neoplasms; Cameroon; Artificial Intelligence; Early Detection of Cancer; Adult; Middle Aged; Qualitative Research; Patient Acceptance of Health Care; Focus Groups
PubMed: 38937771
DOI: 10.1186/s12978-024-01828-8 -
Clinics in Dermatology Jun 2024The last few years have seen a boom in the popularity of artificial intelligence (AI) around the world, and the healthcare sector has not been immune from what has been...
The last few years have seen a boom in the popularity of artificial intelligence (AI) around the world, and the healthcare sector has not been immune from what has been perceived by some as a revolutionary technology. While AI has been around for many years, including in the field of healthcare, the recent introduction of consumer-facing generative AI tools has put a spotlight on the technology that has drawn attention from governments, corporations, consumers and more. Healthcare systems, physician groups, health insurance companies, and others in the space have shown an eagerness to explore AI's potential to improve various aspects of healthcare, but new legal risks and challenges are unfolding every day. This contribution looks at the latest healthcare-related measures in the U.S. and international legal and regulatory landscapes, as well as data privacy implications and discrimination concerns coming out of AI-enabled solutions. It also discusses concerns that healthcare systems and physicians alike are monitoring, including the potential for medical errors resulting from AI, liability considerations, and malpractice insurance trends.
PubMed: 38936641
DOI: 10.1016/j.clindermatol.2024.06.014 -
Journal of Biomedical Informatics Jun 2024Linear and logistic regression are widely used statistical techniques in population genetics for analyzing genetic data and uncovering patterns and associations in large...
OBJECTIVE
Linear and logistic regression are widely used statistical techniques in population genetics for analyzing genetic data and uncovering patterns and associations in large genetic datasets, such as identifying genetic variations linked to specific diseases or traits. However, obtaining statistically significant results from these studies requires large amounts of sensitive genotype and phenotype information from thousands of patients, which raises privacy concerns. Although cryptographic techniques such as homomorphic encryption offers a potential solution to the privacy concerns as it allows computations on encrypted data, previous methods leveraging homomorphic encryption have not addressed the confidentiality of shared models, which can leak information about the training data.
METHODS
In this work, we present a secure model evaluation method for linear and logistic regression using homomorphic encryption for six prediction tasks, where input genotypes, output phenotypes, and model parameters are all encrypted.
RESULTS
Our method ensures no private information leakage during inference and achieves high accuracy (≥93% for all outcomes) with each inference taking less than ten seconds for ∼200 genomes.
CONCLUSION
Our study demonstrates that it is possible to perform linear and logistic regression model evaluation while protecting patient confidentiality with theoretical security guarantees. Our implementation and test data are available at https://github.com/G2Lab/privateML/.
PubMed: 38936565
DOI: 10.1016/j.jbi.2024.104678