Category: Nevin Manimala

PLoS One. 2024 Mar 19;19(3):e0292980. doi: 10.1371/journal.pone.0292980. eCollection 2024.

ABSTRACT

Explainable Artificial Intelligence (XAI) is becoming a disruptive trend in healthcare, allowing for transparency and interpretability of autonomous decision-making. In this study, we present an innovative application of a rule-based classification model to identify the main causes of chronic cough-related quality of life (QoL) impairment in a cohort of asthmatic patients. The proposed approach first involves the design of a suitable symptoms questionnaire and the subsequent analyses via XAI. Specifically, feature ranking, derived from statistically validated decision rules, helped in automatically identifying the main factors influencing an impaired QoL: pharynx/larynx and upper airways when asthma is under control, and asthma itself and digestive trait when asthma is not controlled. Moreover, the obtained if-then rules identified specific thresholds on the symptoms associated to the impaired QoL. These results, by finding priorities among symptoms, may prove helpful in supporting physicians in the choice of the most adequate diagnostic/therapeutic plan.

PMID:38502606 | DOI:10.1371/journal.pone.0292980

AIDS Care. 2024 Mar 19:1-10. doi: 10.1080/09540121.2024.2307383. Online ahead of print.

ABSTRACT

Social network strategy (SNS) testing uses network connections to refer individuals at high risk to HIV testing services (HTS). In Tanzania, SNS testing is offered in communities and health facilities. In communities, SNS testing targets key and vulnerable populations (KVP), while in health facilities it complements index testing by reaching unelicited index contacts. Routine data were used to assess performance and trends over time in PEPFAR-supported sites between October 2021 and March 2023. Key indicators included SNS social contacts tested, and new HIV-positives individuals identified. Descriptive and statistical analysis were conducted. Univariable and multivariable analysis were applied, and variables with P-values <0.2 at univariable analysis were considered for multivariable analysis. Overall, 121,739 SNS contacts were tested, and 7731 (6.4%) previously undiagnosed individuals living with HIV were identified. Tested contacts and identified HIV-positives were mostly aged ≥15 years (>99.7%) and females (80.6% of tests, 79.4% of HIV-positives). Most SNS contacts were tested (78,363; 64.7%) and diagnosed (6376; 82.5%) in communities. SNS tests and HIV-positives grew 11.5 and 6.1-fold respectively, from October-December 2021 to January-March 2023, with majority of clients reached in communities vs. facilities (78,763 vs. 42,976). These results indicate that SNS testing is a promising HIV case-finding approach in Tanzania.

PMID:38502602 | DOI:10.1080/09540121.2024.2307383

Eur J Phys Rehabil Med. 2024 Mar 19. doi: 10.23736/S1973-9087.24.08234-0. Online ahead of print.

ABSTRACT

BACKGROUND: A high rate of hospitalized patients for COVID-19 had dysphagia, frequently underdiagnosed, and not treated, inducing a prolonged dysphagia with protracted recovery. Specific treatments and protocols have not been well described yet.

AIM: Given the potential benefits of respiratory muscle training (IEMT) and neuromuscular stimulation (NMES) in dysphagia treatment, this study aimed to assess the feasibility of the protocol used for treating dysphagia in patients who experienced prolonged hospitalization for COVID-19.

DESIGN: Observational, descriptive, prospective study.

SETTING: Department of Physical Medicine and Rehabilitation of a tertiary University hospital.

POPULATION: Fifty-eight COVID-19 patients were admitted for intensive rehabilitation (March 2020 to October 2021) were prospectively studied.

METHODS: Dysphagia was diagnosed using videofluoroscopy and treated with a 3-week protocol adapted from neuromuscular stimulation (NMES) in a motor threshold and inspiratory/expiratory muscle strength training (IEMST), five sets of five repetitions three times daily for 3 weeks. Feasibility was assessed with adherence, outcomes achieved, and occurrence of adverse/unexpected events. Respiratory function (peak cough flow, maximal inspiratory/expiratory pressures) and swallow function (Penetration-Aspiration Scale and Bolus Residue Scale measured by videofluoroscopy) were recorded descriptive statistics, Student’s t test for numerical data, and Wilcoxon Test for ordinal variables were applied. SPPSS vs28 and STATA version 15.1 (StataCorp, College Station, TX, USA) were used for statistical analysis. P values 0.05 were considered significant.

RESULTS: Dysphagia was highly prevalent in severe COVID-19 patients (86.6%); all respiratory and swallow parameters improved after a 3-week intervention and 12 of 18 patients dependent on tube feeding resumed a normal diet (66.7%; McNemar P=0.03), and 84.09% attended a no restriction diet at discharge. Adherence to treatment was 85%. No significant adverse events were detected.

CONCLUSIONS: We conclude that a structured swallowing-exercise training intervention based on IEMT and NMES is feasible and safe in prolonged hospitalization post-COVID patients.

CLINICAL REHABILITATION IMPACT: To describe rehabilitation protocols used to treat dysphagia in post-COVID patients will help us to optimize the available techniques in each center and to induce a faster recovery avoiding potential complications.

PMID:38502558 | DOI:10.23736/S1973-9087.24.08234-0

Eur J Phys Rehabil Med. 2024 Mar 19. doi: 10.23736/S1973-9087.24.08188-7. Online ahead of print.

ABSTRACT

BACKGROUND: The development of International Classification of Functioning, Disability, and Health (ICF) Core Sets greatly enhances the global recognition of health conditions, thereby advancing research, education, and care provision. Aside from the work of researchers, and the viewpoint of persons with lived experience, the development of Core Sets for deafblindness needs to include the viewpoints of professionals with expertise unique to this condition.

AIM: To represent the perspective of health and social service expert professionals in the development of ICF Core Sets for deafblindness.

DESIGN: Cross-sectional cohort study.

SETTING: Global online survey representing all six regions of the World Health Organization.

POPULATION: One hundred and five professionals providing and health or social service to individuals living with deafblindness with a minimum of 2 years of work experience with this population.

METHODS: An online survey was distributed through professional networks and social media for individuals working with persons living with deafblindness. Demographic items were summarized using descriptive statistics. Six open-ended questions explored the perceptions of body functions and structures that influence activities and participation, as well as environmental and personal factors that facilitate functioning. Data were linked to the ICF codes using established linking rules and procedures.

RESULTS: The 2934 survey response units were linked using IFC categories. Of the 421 unique categories, 133 were used by 5% or more of respondents. Most categories within the Activities and Participation component were equally emphasized. The most frequent Environmental factors were support and relationships, services, systems, and policies, as well as and the physical environment (e.g., hearing aids or noise). Mental functions, including higher level cognitive functions, temperament and personality were frequently emphasized.

CONCLUSIONS: Almost three quarters (73.3%) of the entire ICF classification categories were included in the expert survey results. This proportion emphasizes the importance of a multidimensional tool, such as the ICF, for assessing functioning and health for persons with deafblindness.

CLINICAL REHABILITATION IMPACT: The representation of this professional perspective in Core Set development will improve standardized assessment and documentation, intervention planning, and facilitate interprofessional communication with the goal of improving person-centered care for persons living with deafblindness.

PMID:38502555 | DOI:10.23736/S1973-9087.24.08188-7

Eur J Phys Rehabil Med. 2024 Mar 19. doi: 10.23736/S1973-9087.24.08177-2. Online ahead of print.

ABSTRACT

BACKGROUND: The main treatment aim in mild scoliosis is to prevent progression and if possible, to avoid bracing. Physiotherapeutic Scoliosis Specific Exercises (PSSE) are curve pattern specific exercises, based on 3D self-correction and activities of daily living training.

AIM: The objective of this study was to evaluate the efficacy of PSSE – Schroth, as an exclusive treatment, during the riskiest period of rapid growth.

DESIGN: Prospective control study.

SETTING: Outpatient treatment.

POPULATION: Adolescents with scoliosis.

METHODS: One hundred and sixty-three patients (148 girls,15 boys; mean age 12.6 years, Risser sign 1.1, thoracic (Th) Cobb angle 20.8° and lumbar/thoracolumbar (L/TL) Cobb angle 20.7°) performed PSSE – Schroth exercises in our clinic. They were asked to regularly attend supervised sessions and to follow a home-program at least 5 times per week. Our inclusion criteria were Cobb angle 15°-25°, Risser 0-2 and angle trunk rotation (ATR) >5°, measured by scoliometer. The outcome parameters were the Cobb angle before and after the intervention (improvement or progression were defined as angle difference more than 5°) and the number of patients that finally needed a brace. Average follow up time was 29.4 months. Control group was consisted of 58 patients (54 girls, 4 boys; mean age 13.1 years, Risser sign 0-2, Th Cobb 19.4°, L/TL Cobb 19.2°), that were retrospectively analyzed and performed general or no exercises. Compliance was self-reported. Statistical analysis was performed by paired t-test.

RESULTS: For PSSE – Schroth group, 103 patients (63.2%) remained stable, 39 (23.9%) improved and 21 (12.9%) worsened. The success rate (87.1%) was significantly higher compared to Control group (P=0.002), where 15 subjects (25.9%) were stable and 43 (74.1%) worsened. Similarly, 16 patients (9.8%) from PSSE – Schroth group finally needed a brace, while 39 (67.2%) from control group (P=0.01).

CONCLUSIONS: PSSE – Schroth reduced the risk of progression in Adolescent Idiopathic Scoliosis (AIS) patients, during early growth. Our results are in accordance with the recently published literature, showing the effectiveness of PSSE and their superiority compared to general exercises or natural history.

CLINICAL REHABILITATION IMPACT: Scoliosis specific exercises can be the first step of scoliosis treatment in mild curves, to avoid progression and bracing.

PMID:38502554 | DOI:10.23736/S1973-9087.24.08177-2

JMIR Infodemiology. 2024 Mar 19;4:e51113. doi: 10.2196/51113.

ABSTRACT

BACKGROUND: The COVID-19 pandemic triggered unprecedented global vaccination efforts, with social media being a popular tool for vaccine promotion.

OBJECTIVE: This study probes into Macao’s COVID-19 vaccine communication dynamics, with a focus on the multifaceted impacts of government agendas on social media.

METHODS: We scrutinized 22,986 vaccine-related Facebook posts from January 2020 to August 2022 in Macao. Using automated content analysis and advanced statistical methods, we unveiled intricate agenda dynamics between government and nongovernment entities.

RESULTS: “Vaccine importance” and “COVID-19 risk” were the most prominent topics co-occurring in the overall vaccine communication. The government tended to emphasize “COVID-19 risk” and “vaccine effectiveness,” while regular users prioritized vaccine safety and distribution, indicating a discrepancy in these agendas. Nonetheless, the government has limited impact on regular users in the aspects of vaccine importance, accessibility, affordability, and trust in experts. The agendas of government and nongovernment users intertwined, illustrating complex interactions.

CONCLUSIONS: This study reveals the influence of government agendas on public discourse, impacting environmental awareness, public health education, and the social dynamics of inclusive communication during health crises. Inclusive strategies, accommodating public concerns, and involving diverse stakeholders are paramount for effective social media communication during health crises.

PMID:38502184 | DOI:10.2196/51113

JMIR Public Health Surveill. 2024 Mar 19;10:e48738. doi: 10.2196/48738.

ABSTRACT

BACKGROUND: Population size, prevalence, and incidence are essential metrics that influence public health programming and policy. However, stakeholders are frequently tasked with setting performance targets, reporting global indicators, and designing policies based on multiple (often incongruous) estimates of these variables, and they often do so in the absence of a formal, transparent framework for reaching a consensus estimate.

OBJECTIVE: This study aims to describe a model to synthesize multiple study estimates while incorporating stakeholder knowledge, introduce an R Shiny app to implement the model, and demonstrate the model and app using real data.

METHODS: In this study, we developed a Bayesian hierarchical model to synthesize multiple study estimates that allow the user to incorporate the quality of each estimate as a confidence score. The model was implemented as a user-friendly R Shiny app aimed at practitioners of population size estimation. The underlying Bayesian model was programmed in Stan for efficient sampling and computation.

RESULTS: The app was demonstrated using biobehavioral survey-based population size estimates (and accompanying confidence scores) of female sex workers and men who have sex with men from 3 survey locations in a country in sub-Saharan Africa. The consensus results incorporating confidence scores are compared with the case where they are absent, and the results with confidence scores are shown to perform better according to an app-supplied metric for unaccounted-for variation.

CONCLUSIONS: The utility of the triangulator model, including the incorporation of confidence scores, as a user-friendly app is demonstrated using a use case example. Our results offer empirical evidence of the model’s effectiveness in producing an accurate consensus estimate and emphasize the significant impact that the accessible model and app offer for public health. It offers a solution to the long-standing problem of synthesizing multiple estimates, potentially leading to more informed and evidence-based decision-making processes. The Triangulator has broad utility and flexibility to be adapted and used in various other contexts and regions to address similar challenges.

PMID:38502183 | DOI:10.2196/48738

JMIR Cancer. 2024 Mar 19;10:e52322. doi: 10.2196/52322.

ABSTRACT

BACKGROUND: People with cancer frequently experience severe and distressing symptoms associated with cancer and its treatments. Predicting symptoms in patients with cancer continues to be a significant challenge for both clinicians and researchers. The rapid evolution of machine learning (ML) highlights the need for a current systematic review to improve cancer symptom prediction.

OBJECTIVE: This systematic review aims to synthesize the literature that has used ML algorithms to predict the development of cancer symptoms and to identify the predictors of these symptoms. This is essential for integrating new developments and identifying gaps in existing literature.

METHODS: We conducted this systematic review in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist. We conducted a systematic search of CINAHL, Embase, and PubMed for English records published from 1984 to August 11, 2023, using the following search terms: cancer, neoplasm, specific symptoms, neural networks, machine learning, specific algorithm names, and deep learning. All records that met the eligibility criteria were individually reviewed by 2 coauthors, and key findings were extracted and synthesized. We focused on studies using ML algorithms to predict cancer symptoms, excluding nonhuman research, technical reports, reviews, book chapters, conference proceedings, and inaccessible full texts.

RESULTS: A total of 42 studies were included, the majority of which were published after 2017. Most studies were conducted in North America (18/42, 43%) and Asia (16/42, 38%). The sample sizes in most studies (27/42, 64%) typically ranged from 100 to 1000 participants. The most prevalent category of algorithms was supervised ML, accounting for 39 (93%) of the 42 studies. Each of the methods-deep learning, ensemble classifiers, and unsupervised ML-constituted 3 (3%) of the 42 studies. The ML algorithms with the best performance were logistic regression (9/42, 17%), random forest (7/42, 13%), artificial neural networks (5/42, 9%), and decision trees (5/42, 9%). The most commonly included primary cancer sites were the head and neck (9/42, 22%) and breast (8/42, 19%), with 17 (41%) of the 42 studies not specifying the site. The most frequently studied symptoms were xerostomia (9/42, 14%), depression (8/42, 13%), pain (8/42, 13%), and fatigue (6/42, 10%). The significant predictors were age, gender, treatment type, treatment number, cancer site, cancer stage, chemotherapy, radiotherapy, chronic diseases, comorbidities, physical factors, and psychological factors.

CONCLUSIONS: This review outlines the algorithms used for predicting symptoms in individuals with cancer. Given the diversity of symptoms people with cancer experience, analytic approaches that can handle complex and nonlinear relationships are critical. This knowledge can pave the way for crafting algorithms tailored to a specific symptom. In addition, to improve prediction precision, future research should compare cutting-edge ML strategies such as deep learning and ensemble methods with traditional statistical models.

PMID:38502171 | DOI:10.2196/52322

Rev Neurol. 2024 Apr 1;78(7):209-211. doi: 10.33588/rn.7807.2023164.

ABSTRACT

Leading scientific journals in fields such as medicine, biology and sociology repeatedly publish articles and editorials claiming that a large percentage of doctors do not understand the basics of statistical analysis, which increases the risk of errors in interpreting data, makes them more vulnerable to misinformation and reduces the effectiveness of research. This problem extends throughout their careers and is largely due to the poor training they receive in statistics – a problem that is common in developed countries. As stated by H. Halle and S. Krauss, ‘90% of German university lecturers who regularly use the p-value in tests do not understand what that value actually measures’. It is important to note that the basic reasoning of statistical analysis is similar to what we do in our daily lives and that understanding the basic concepts of statistical analysis does not require any knowledge of mathematics. Contrary to what many researchers believe, the p-value of the test is not a ‘mathematical index’ that allows us to clearly conclude whether, for example, a drug is more effective than a placebo. The p-value of the test is simply a percentage.

PMID:38502169 | DOI:10.33588/rn.7807.2023164

JMIR Form Res. 2024 Mar 19;8:e55041. doi: 10.2196/55041.

ABSTRACT

BACKGROUND: Alcohol misuse is the fourth leading cause of death in the United States and a significant problem in the US military. Brief alcohol interventions can reduce negative alcohol outcomes in civilian and military populations, but additional scalable interventions are needed to reduce binge and heavy drinking. SMS text messaging interventions could address this need, but to date, no programs exist for military populations.

OBJECTIVE: We aimed to develop an SMS text messaging intervention to address binge and heavy drinking among Airmen in Technical Training in the US Air Force.

METHODS: We implemented a 2-phase, mixed methods study to develop the SMS text messaging intervention. In phase 1, a total of 149 respondents provided feedback about the persuasiveness of 49 expert-developed messages, preferences regarding message frequency, timing and days to receive messages, and suggested messages, which were qualitatively coded. In phase 2, a total of 283 respondents provided feedback about the persuasiveness of 77 new messages, including those developed through the refinement of messages from phase 1, which were coded and assessed based on the Behavior Change Technique Taxonomy (BCTT). For both phases, mean persuasiveness scores (range 1-5) were calculated and compared according to age (aged <21 or ≥21 years) and gender. Top-ranking messages from phase 2 were considered for inclusion in the final message library.

RESULTS: In phase 1, top-rated message themes were about warnings about adverse outcomes (eg, impaired judgment and financial costs), recommendations to reduce drinking, and invoking values and goals. Through qualitative coding of suggested messages, we identified themes related to warnings about adverse outcomes, recommendations, prioritizing long-term goals, team and belonging, and invoking values and goals. Respondents preferred to receive 1 to 3 messages per week (124/137, 90.5%) and to be sent messages on Friday, Saturday, and Sunday (65/142, 45.8%). In phase 2, mean scores for messages in the final message library ranged from 3.31 (SD 1.29) to 4.21 (SD 0.90). Of the top 5 highest-rated messages, 4 were categorized into 2 behavior change techniques (BCTs): valued self-identity and information about health consequences. The final message library includes 28 BCTT-informed messages across 13 BCTs, with messages having similar scores across genders. More than one-fourth (8/28, 29%) of the final messages were informed by the suggested messages from phase 1. As Airmen aged <21 years face harsher disciplinary action for alcohol consumption, the program is tailored based on the US legal drinking age.

CONCLUSIONS: This study involved members from the target population throughout 2 formative stages of intervention development to design a BCTT-informed SMS text messaging intervention to reduce binge and heavy drinking, which is now being tested in an efficacy trial. The results will determine the impact of the intervention on binge drinking and alcohol consumption in the US Air Force.

PMID:38502165 | DOI:10.2196/55041