Clinical eHealth最新文献

In the realm of intensive care medicine, wearable electronic devices have emerged as a highly promising field, driven by advancements in mobile, intelligent, and personalized healthcare. They are defined as devices that can be worn directly on the body, offering portable services by actively recording physiological parameters and metabolic status, providing index monitoring, clinical diagnosis, and disease treatment. This review specifically highlights the utilization of wearable devices in intensive care units within the field of intensive care medicine, anticipating their future applications.

This research work presents a hybrid approach combining a type-2 fuzzy inference system with particle swarm optimization (PSO) to develop a type-2 fuzzy optimized inference system, specifically tailored for asthma patient data. Addressing the inherent uncertainty in medical diagnostics, this model enhances traditional type-1 fuzzy logic by incorporating ambiguity into linguistic variables and utilizing type-2 fuzzy if-then rules. The system is trained to minimize diagnostic error in asthma disease identification. Applied to a dataset comprising eight medical entities from asthma patients, the model demonstrates substantial accuracy improvements. Numerical computations validate the system, showing a decrease in error rate from 1.445 to 0.03, indicating a significant enhancement in diagnostic precision. These results underscore the potential of our model in medical diagnostic problems, providing a novel and effective tool for tackling the complexities of asthma diagnosis.

The existing global health crisis characterized by limited resources, including health personnel, has prompted the adoption of telemedicine and telehealth, especially in the post-pandemic era. The COVID-19 pandemic accelerated the adoption of these technologies, and as the world navigates beyond the crisis, it is essential to assess the extent of utilization of telehealth and telemedicine. This review study aims to assess the extent to which teleservices have been implemented worldwide across different continents. Peer-reviewed telehealth and telemedicine articles were reviewed across Web of Science, Scopus, Cochrane Library and Pubmed/Medline databases. The exclusion criteria comprised all articles published before 1 December 2019, any other databases, duplicates, and grey literature. The inclusion criteria for this study encompassed articles published on or after December 1, 2019. This timeframe allowed us to focus on the pandemic and post-pandemic era. A total of 381 publications were identified for inclusion, which were screened based on reviewing the titles, abstracts and full-text content down to 102 relevant articles. Utilization trends were identified amongst the different countries across the continents, and these were classified into advanced, developed, developing and emerging adoption stages. The respective characteristics utilized in differentiating these adoption stages were identified, encompassing the inclusivity of teleservices in administration, disease diagnosis, treatment, patient follow-ups, pharmacy services and electronic health records transversely. According to this review, of the countries surveyed (n = 77), (n = 27) 13.8 % are at an advanced adoption stage, (n = 20) 10.3 % are at a developed level, (n = 24) 12.3 % are at the developed stage, and (n = 6) 3 % are at the emerging stage, as percentages of all the countries in the world (N = 195). In conclusion, this study demonstrated the extent to which various nations have adopted telehealth and telemedicine from the onset of the COVID-19 pandemic to 2023.

Medical Internet of Things (MIoT) and Digital healthcare have long since ceased to be separate entities. Cross-fertilisation of MIoT-based digital medical treatment models is the way forward. However, few studies discuss the development of MIoT and Digital healthcare in clinical practice. This paper reviews the research progress of MIoT and Digital healthcare in related fields.

Lung cancer is the leading cause of cancer-related deaths worldwide. Early diagnosis and treatment play a crucial role in improving the prognosis for lung cancer. However, the issue of overtreatment and delayed diagnosis remains prevalent due to the considerable limitations of manual film review in facilitating early detection and treatment of lung cancer. In recent years, artificial intelligence (AI) has emerged as a valuable tool for clinicians to screen and evaluate benign and malignant pulmonary nodules, offering numerous advantages. Nevertheless, the sensitivity and specificity of AI are neither sufficient to completely replace medical experts nor capable of assuming direct responsibility for clinical diagnosis and treatment.

Therefore, we propose the concept of a Human-Computer Multi-Disciplinary Team (MDT), which involves collaborative decision-making between human physicians and AI systems. The human-computer MDT approach in pulmonary nodule evaluation presents a novel model for diagnosis and treatment, leveraging the respective strengths of human expertise and AI capabilities. This review provides an overview of the background, medical application, advantages and limitations, future trends, and reporting format of the Human-Computer MDT in pulmonary nodule evaluation.

Its aim is to explore standardized methods for enhancing early diagnosis in lung cancer. With the rapid advancement of AI and the field of meta-cosmic medicine, human-computer MDT are expected to become more widespread and play an important role in the implementation of the Healthy China 2030 plan, particularly in improving primary medical care in the future.

For early-stage non-small cell lung cancer, surgical resection was used as the first treatment. However, approximately 20% of patients were not suitable for surgery due to severe comorbidities. We verified the feasibility, safety, and tolerability of percutaneous thermal ablation for patients of chronic obstructive pulmonary disease (COPD) with peripheral high-risk pulmonary nodules in the real world. The patients with peripheral high-risk pulmonary nodules ineligible or unwilling to undergo surgery who were ineligible or unwilling tosurgery underwent CT-guided thermal ablation in our hospital from January 1st, 2019 to May 31th, 2022 were retrospectively collected, and divided into COPD and non COPD group. Incidence, severity, risk factors of complications between in different severity of COPD and non-COPD group were compared. A total of 216 high-risk were enrolled, including 73 in COPD group and 143 in the non-COPD group. The average age, male gender, MMRC score, size of nodules, incidence of confirmed pathological diagnosis, and pneumothorax after thermal ablation were higher in the COPD group than in the non-COPD group. COPD was the only independent risk factor for pneumothorax after ablation. The incidence of pneumothorax increased with the severity of COPD, but no statistical significance. Compared to the baseline, the MMRC score was significantly increased in the COPD group, but there was no significant difference in the discharge time and hospitalization expenses between the COPD patients with or without pneumothorax. CT-guided percutaneous thermal ablation is a safe and feasible therapy for different severities of COPD with high-risk pulmonary nodules, and it is well-tolerated without increasing medical burden.

Parkinson’s disease (PD) is a devastating neurological syndrome that affects millions of people worldwide. For the successful treatment and control of PD, it is essential to detect it early and diagnose it accurately. Machine learning (ML) algorithms have shown promising results in identifying PD based on various clinical and non-invasive measures. This paper proposes an ensemble classifier-based method to identify PD using ML algorithms. We consider two classes of PD, namely, healthy controls and PD patients. Our approach involves the use of feature selection, feature extraction, and classification techniques to develop a robust and accurate model. We use a dataset that includes clinical measures and necessary features from patients with PD and healthy controls. Our outcomes demonstrate the effectiveness of the proposed method in accurately identifying PD and highlight the importance of ML algorithms in assisting with early detection and diagnosis of PD.