Identification of Patient Profiles with High Risk of Hospital Re-Admissions for Acute COPD Exacerbations (AECOPD) in France Using a Machine Learning Model

Long-term inhaled corticosteroid treatment in patients with chronic obstructive pulmonary disease, cardiovascular disease, and a recent hospitalised exacerbation: The ICSLIFE pragmatic, randomised controlled study

INTRODUCTION

Patients with chronic obstructive pulmonary disease (COPD) frequently have cardiovascular comorbidities, increasing the risk of hospitalised COPD exacerbations (H-ECOPDs) or death. This pragmatic study examined the effects of adding an inhaled corticosteroid (ICS) to long-acting bronchodilator(s) (LABDs) in patients with COPD and cardiac comorbidities who had a recent H-ECOPD.

METHODS

Patients >60 years of age with COPD and ≥1 cardiac comorbidity, within 6 months after discharge following an H-ECOPD, were randomised to receive LABD(s) with or without ICS, and were followed for 1 year. The primary outcome was the time to first rehospitalisation and/or all-cause death.

RESULTS

The planned number of patients was not recruited (803/1032), limiting the strength of the conclusions. In the intention-to-treat population, 89/403 patients (22.1 %) were rehospitalised or died in the LABD group (probability 0.257 [95 % confidence interval 0.206, 0.318]), vs 85/400 (21.3 %) in the LABD+ICS group (0.249 [0.198, 0.310]), with no difference between groups in time-to-event (hazard ratio 1.116 [0.827, 1.504]; p = 0.473). All-cause and cardiovascular mortality were lower in patients receiving LABD(s)+ICS, with relative reductions of 19.7 % and 27.4 %, respectively (9.8 % vs 12.2 % and 4.5 % vs 6.2 %), although the groups were not formally statistically compared for these endpoints. Fewer patients had adverse events in the LABD+ICS group (43.0 % vs 50.4 %; p = 0.013), with 4.9 % vs 5.4 % reporting pneumonia adverse events.

CONCLUSIONS

Results suggest addition of ICS to LABDs did not reduce the time-to-combined rehospitalisation/death, although it decreased all-cause and cardiovascular mortality. ICS use was not associated with an increased risk of adverse events, particularly pneumonia.

Impact of the COVID-19 Pandemic on COPD Patient Mortality: A Nationwide Study in France

Objectives: We investigated the mortality patterns of chronic obstructive pulmonary disease (COPD) patients in France relative to a control population, comparing year 2020 to pre-pandemic years 2017-2019. Methods: COPD patient and sex, age and residence matched control cohorts were created from the French National Health Data System. Survival was analyzed using Cox regressions and standardized rates. Results: All-cause mortality increased in 2020 compared to 2019 in the COPD population (+4%), but to a lesser extent than in the control population (+10%). Non-COVID-19 mortality decreased to a greater extent in COPD patients (-5%) than in the controls (-2%). Death rate from COVID-19 was twice as high in the COPD population relative to the control population (547 vs. 279 per 100,000 person-years). Conclusion: The direct impact of the pandemic in terms of deaths from COVID-19 was much greater in the COPD population than in the control population. However, the larger decline in non-COVID-19 mortality in COPD patients could reflect a specific protective effect of the containment measures on this population, counterbalancing the direct impact they had been experiencing.

Demystification of artificial intelligence for respiratory clinicians managing patients with obstructive lung diseases

INTRODUCTION

Asthma and chronic obstructive pulmonary disease (COPD) are leading causes of morbidity and mortality worldwide. Despite all available diagnostics and treatments, these conditions pose a significant individual, economic and social burden. Artificial intelligence (AI) promises to support clinical decision-making processes by optimizing diagnosis and treatment strategies of these heterogeneous and complex chronic respiratory diseases. Its capabilities extend to predicting exacerbation risk, disease progression and mortality, providing healthcare professionals with valuable insights for more effective care. Nevertheless, the knowledge gap between respiratory clinicians and data scientists remains a major constraint for wide application of AI and may hinder future progress. This narrative review aims to bridge this gap and encourage AI deployment by explaining its methodology and added value in asthma and COPD diagnosis and treatment.

AREAS COVERED

This review offers an overview of the fundamental concepts of AI and machine learning, outlines the key steps in building a model, provides examples of their applicability in asthma and COPD care, and discusses barriers to their implementation.

EXPERT OPINION

Machine learning can advance our understanding of asthma and COPD, enabling personalized therapy and better outcomes. Further research and validation are needed to ensure the development of clinically meaningful and generalizable models.

Correlation of Serum Clara Cell Secretory Protein 16, Plasma Fibrinogen and Serum Amyloid A with the Severity of Acute Exacerbated COPD and Their Combination in Prognosis Assessment

Introduction

Chronic obstructive pulmonary disease (COPD) has tremendous detrimental effects on patients' quality of life, lung function, disease progression and socioeconomic burden. This study aimed to investigate new serum biomarkers for COPD detection. Three recently emerging biomarkers, including Clara cell secretory protein⁃16 (CC16), plasma fibrinogen (FIB) and serum amyloid A (SAA), were investigated for their potential in stratifying the severity of COPD.

Methods

A total of 220 patients with AECOPD were recruited. Multivariate logistical regression was used to analyze odds ratios of an array of characteristic of patients, including age, global initiative for chronic obstructive lung disease (GOLD), diabetes mellitus, heart diseases, PaCO2, CC16, FIB, and SAA. Correlations of CC16, FIB and SAA levels to each other, GOLD, and PaCO2 were also measured using Spearman correlation. Receiver operating characteristic (ROC)/curve analysis was used to assess sensitivity and specificity of CC16, FIB, SAA and the combination of the three markers in identifying AECOPD patients with poor prognosis.

Results

Our data suggested that age, GOLD, diabetes mellitus, heart diseases, PaCO2, CC16, FIB, and SAA are all significant risk factors for poor prognosis of AECOPD. CC16, FIB and SAA were positively correlated to each other and to GOLD and PaCO2 levels. CC16, FIB and SAA all had a high sensitivity and specificity in identifying patients with a poor prognosis. CC16, FIB and SAA are new markers with potentially high predictive value in AECOPD.

Discussion

Our data support further development of these biomarkers to improve clinical management of AECOPD through providing more accurate prognosis of AECOPD patients that enable timely adjustment of treatment plans.

Incidence, risk factors, and burden of incisional hernia repair after abdominal surgery in France: a nationwide study

PURPOSE

Incisional hernias are common after laparotomies. The aims of this study were to assess the rate of incisional hernia repair after abdominal surgery, recurrence rate, hospital costs, and risk factors, in France.

METHODS

This national, retrospective, longitudinal, observational study was based on the exhaustive hospital discharge database (PMSI). All adult patients (≥ 18 years old) hospitalised for an abdominal surgical procedure between 01-01-2013 and 31-12-2014 and hospitalised for incisional hernia repair within five years were included. Descriptive analyses and cost analyses from the National Health Insurance (NHI) viewpoint (hospital care for the hernia repair) were performed. To identify risk factors for hernia repair a multivariable Cox model and a machine learning analysis were performed.

RESULTS

In 2013-2014, 710074 patients underwent abdominal surgery, of which 32633 (4.6%) and 5117 (0.7%) had ≥ 1 and ≥ 2 incisional hernia repair(s) within five years, respectively. Mean hospital costs amounted to €4153/hernia repair, representing nearly €67.7 million/year. Some surgical sites exposed patients at high risk of incisional hernia repair: colon and rectum (hazard ratio [HR] 1.2), and other sites on the small bowel and the peritoneum (HR 1.4). Laparotomy procedure and being ≥ 40 years old put patients at high risk of incisional hernia repair even when operated on low-risk sites such as stomach, duodenum, and hepatobiliary.

CONCLUSION

The burden of incisional hernia repair is high and most patients are at risk either due to age ≥ 40 or the surgery site. New approaches to prevent the onset of incisional hernia are warranted.

Readmission rate for acute exacerbation of chronic obstructive pulmonary disease: A systematic review and meta-analysis

BACKGROUND

The readmission rate following hospitalization for chronic obstructive pulmonary disease (COPD) exacerbations is extremely high and has become a common and challenging clinical problem. This study aimed to systematically summarize COPD readmission rates for acute exacerbations and their underlying risk factors.

METHODS

A comprehensive search was performed using PubMed, Embase, Cochrane Library, and Web of Science, published from database inception to April 2, 2022. Methodological quality was evaluated using the Newcastle-Ottawa Scale (NOS). We used a random-effects model or a fixed-effects model to estimate the pooled COPD readmission rate for acute exacerbations and underlying risk factors.

RESULTS

A total of 46 studies were included, of which 24, 7, 17, 7, and 20 summarized the COPD readmission rates for acute exacerbations within 30, 60, 90, 180, and 365 days, respectively. The pooled 30-, 60-, 90-, 180-, and 365-day readmission rates were 11%, 17%, 17%, 30%, and 37%, respectively. The study design type, age stage, WHO region, and length of stay (LOS) were initially considered to be sources of heterogeneity. We also identified potential risk factors for COPD readmission, including male sex, number of hospitalizations in the previous year, LOS, and comorbidities such as heart failure, tumor or cancer, and diabetes, whereas obesity was a protective factor.

CONCLUSIONS

Patients with COPD had a high readmission rate for acute exacerbations, and potential risk factors were identified. Therefore, we should propose clinical interventions and adjust or targeted the control of avoidable risk factors to prevent and reduce the negative impact of COPD readmission.

SYSTEMATIC REVIEW REGISTRATION

PROSPERO, identifier CRD42022333581.

Explaining predictive factors in patient pathways using autoencoders

This paper introduces an end-to-end methodology to predict a pathway-related outcome and identifying predictive factors using autoencoders. A formal description of autoencoders for explainable binary predictions is presented, along with two objective functions that allows for filtering and inverting negative examples during training. A methodology to model and transform complex medical event logs is also proposed, which keeps the pathway information in terms of events and time, as well as the hierarchy information carried in medical codes. A case study is presented, in which the short-term mortality after the implementation of an Implantable Cardioverter-Defibrillator is predicted. Proposed methodologies have been tested and compared to other predictive methods, both explainable and not explainable. Results show the competitiveness of the method in terms of performances, particularly the use of a Variational Auto Encoder with an inverse objective function. Finally, the explainability of the method has been demonstrated, allowing for the identification of interesting predictive factors validated using relative risks.

Hospital admissions and mortality for acute exacerbations of COPD during the COVID-19 pandemic: A nationwide study in France

Background

A global reduction in hospital admissions for acute exacerbations of chronic obstructive pulmonary disease (AECOPD) was observed during the first months of the COVID-19 pandemic. Large-scale studies covering the entire pandemic period are lacking. We investigated hospitalizations for AECOPD and the associated in-hospital mortality at the national level in France during the first 2 years of the pandemic.

Methods

We used the French National Hospital Database to analyse the time trends in (1) monthly incidences of hospitalizations for AECOPD, considering intensive care unit (ICU) admission and COVID-19 diagnoses, and (2) the related in-hospital mortality, from January 2016 to November 2021. Pandemic years were compared with the pre-pandemic years using Poisson regressions.

Results

The database included 565,890 hospitalizations for AECOPD during the study period. The median age at admission was 74 years (interquartile range 65–83), and 37% of the stays concerned women. We found: (1) a dramatic and sustainable decline in hospitalizations for AECOPD over the pandemic period (from 8,899 to 6,032 monthly admissions, relative risk (RR) 0.65, 95% confidence interval (CI) 0.65–0.66), and (2) a concomitant increase in in-hospital mortality for AECOPD stays (from 6.2 to 7.6% per month, RR 1.24, 95% CI 1.21–1.27). The proportion of stays yielding ICU admission was similar in the pre-pandemic and pandemic years, 21.5 and 21.3%, respectively. In-hospital mortality increased to a greater extent for stays without ICU admission (RR 1.39, 95% CI 1.35–1.43) than for those with ICU admission (RR 1.09, 95% CI 1.05–1.13). Since January 2020, only 1.5% of stays were associated with a diagnosis of COVID-19, and their mortality rate was nearly three-times higher than those without COVID-19 (RR 2.66, 95% CI 2.41–2.93).

Conclusion

The decline in admissions for AECOPD during the pandemic could be attributed to a decrease in the incidence of exacerbations for COPD patients and/or to a possible shift from hospital to community care. The rise in in-hospital mortality is partially explained by COVID-19, and could be related to restricted access to ICUs for some patients and/or to greater proportions of severe cases among the patients hospitalized during the pandemic.

Evaluation of a patient-centered integrated care program for individuals with frequent hospital readmissions and multimorbidity

Managing patients with multimorbidity and frequent hospital readmissions is a challenge. Integrated care programs that consider their needs and allow for personalized care are necessary for their early identification and management. This work aims to describe these patients' clinical characteristics and evaluate a program designed to reducing readmissions. This prospective study analyzed all patients with ≥ 3 admissions to a medical department in the previous year who were included in the Internal Medicine Department chronic care program at the Lucus Augusti University Hospital (Lugo, Spain) between April 1, 2019 and April 30, 2021. A multidimensional assessment, personalized care plan, and proactive follow-up with a case manager nurse were provided via an advanced hospital system. Clinical and demographic variables and data on healthcare system use were analyzed at 6 and 12 months before and after inclusion. Descriptive and survival analyses were performed. One hundred sixty-one patients were included. Program participants were elderly (mean 81.4 (SD 11) years), had multimorbidity (10.2 (3) chronic diseases) and polypharmacy (10.6 (3.5) drugs), frequently used the healthcare system, and were highly complex. Most were included for heart failure. The program led to significant reductions in admissions and emergency department visits (p = .0001). A total of 44.7% patients died within 1 year. The PROFUND Index showed good predictive ability (p = .013), with high values associated with mortality (RR 1.15, p = .001). Patients with frequent hospital readmissions are highly complex and need special care. A personalized integrated care program reduced admissions and allowed for individualized decision-making.

Bone Marrow Mesenchymal Stem Cell Transplantation in Combination with Nasal Continuous Positive Airway Pressure Improves Acute Exacerbation of Chronic Obstructive Pulmonary Disease (AECOPD)

We aimed to explore the efficacy of bone marrow mesenchymal stem cell (BMSC) transplantation combined with nasal continuous positive airway pressure (nCPAP) for treating severe acute exacerbation of chronic obstructive pulmonary disease (AECOPD). SD rat AECOPD model was established by injecting endotoxin and Staphylococcus aureus and then treated with nCPAP, BMSCs, or nCPAP combined with BMSCs (n = 20) and their conditions were evaluated with BBB score at 1 d, 3 d, 7 d, 14 d, 28 d after treatment along with analysis of apoptosis and BrdU-positive cells as well as NF200 expression by TUNEL kit staining and levels of Th1, Th7 and Th12 before and after treatment. As revealed by BBB score and HE staining, all treatments significantly alleviated the symptom of severe APEOPD (p < 0.05), while compared with nCPAP, the combined treatment exhibited higher efficacy. Besides, upon treatment, apoptosis and level of Th1, Th7 and Th12 was reduced but N200 absorbance value was elevated, with significant difference in combination group (p < 0.05). In conclusion, BMSC transplantation in combination with nCPAP alleviates severe AECOPD by reducing cell apoptosis, repairing cell damage, and regulating T-cell subsets.

Benefit of Prompt versus Delayed Use of Single-Inhaler Fluticasone Furoate/Umeclidinium/Vilanterol (FF/UMEC/VI) Following a COPD Exacerbation

Purpose

Triple therapy (TT; inhaled corticosteroid, long-acting muscarinic antagonist, and long-acting β₂-agonist) is recommended for patients with chronic obstructive pulmonary disease (COPD) at risk of exacerbation, although the optimum timing of TT initiation remains unclear. This study evaluated the impact of prompt versus delayed initiation of single-inhaler TT (fluticasone furoate, umeclidinium, and vilanterol [FF/UMEC/VI]) following a COPD exacerbation.

Patients and Methods

This retrospective cohort study used data from the IQVIA PharMetrics^® Plus database. Patients initiating FF/UMEC/VI following a COPD exacerbation between September 18, 2017 and September 30, 2019 (exacerbation = index date) were categorized as prompt (within 30 days of index) or delayed (31–180 days after index) FF/UMEC/VI initiators. Patients were aged ≥40 years at index, had ≥12 months’ continuous health insurance coverage before index (baseline), and ≥6 months’ coverage after index (follow-up). Patients with a COPD exacerbation or claim for FF/UMEC/VI during baseline were excluded. Inverse probability weighting was used to adjust for differences in baseline characteristics between cohorts. Exacerbations (overall, moderate, and severe), healthcare costs, and readmissions were evaluated during follow-up.

Results

A total of 1904 patients (prompt: 529; delayed: 1375) were included. After weighting, baseline characteristics were well balanced between cohorts. Patients in the prompt cohort had significantly lower rates per person-year (PPY) of overall (0.98 vs 1.23; rate ratio [RR] [95% CI] = 0.79 [0.65–0.94], p = 0.004), moderate (0.86 vs 1.03; RR [95% CI] = 0.84 [0.69–0.99], p = 0.038), and severe (0.11 vs 0.20; RR [95% CI] = 0.57 [0.37–0.79], p = 0.002) exacerbations, compared with delayed initiators. Mean all-cause and COPD-related healthcare costs were significantly lower among prompt initiators (all-cause: $26,107 vs $32,400 PPY, p = 0.014; COPD-related: $12,694 vs $17,640 PPY, p = 0.002).

Conclusion

Prompt initiation of FF/UMEC/VI following a moderate or severe COPD exacerbation was associated with significant reductions in exacerbations and healthcare costs relative to delayed initiation.

Digital healthcare in COPD management: a narrative review on the advantages, pitfalls, and need for further research

Chronic obstructive pulmonary disease (COPD) remains a leading cause of morbidity and mortality despite current treatment strategies which focus on smoking cessation, pulmonary rehabilitation, and symptomatic relief. A focus of COPD care is to encourage self-management, particularly during COVID-19, where much face-to-face care has been reduced or ceased. Digital health solutions may offer affordable and scalable solutions to support COPD patient education and self-management, such solutions could improve clinical outcomes and expand service reach for limited additional cost. However, optimal ways to deliver digital medicine are still in development, and there are a number of important considerations for clinicians, commissioners, and patients to ensure successful implementation of digitally augmented care. In this narrative review, we discuss advantages, pitfalls, and future prospects of digital healthcare, which offer a variety of tools including self-management plans, education videos, inhaler training videos, feedback to patients and healthcare professionals (HCPs), exacerbation monitoring, and pulmonary rehabilitation. We discuss the key issues with sustaining patient and HCP engagement and limiting attrition of use, interoperability with devices, integration into healthcare systems, and ensuring inclusivity and accessibility. We explore the essential areas of research beyond determining safety and efficacy to understand the acceptability of digital healthcare solutions to patients, clinicians, and healthcare systems, and hence ways to improve this and sustain engagement. Finally, we explore the regulatory challenges to ensure quality and engagement and effective integration into current healthcare systems and care pathways, while maintaining patients' autonomy and privacy. Understanding and addressing these issues and successful incorporation of an acceptable, simple, scalable, affordable, and future-proof digital solution into healthcare systems could help remodel global chronic disease management and fractured healthcare systems to provide best patient care and optimisation of healthcare resources to meet the global burden and unmet clinical need of COPD.

Costs and mortality associated with HIV: a machine learning analysis of the French national health insurance database

BACKGROUND

The objective is to characterise the economic burden to the healthcare system of people living with HIV (PLWHIV) in France and to help decision makers in identifying risk factors associated with high-cost and high mortality profiles.

DESIGN AND METHODS

The study is a retrospective analysis of PLWHIV identified in the French National Health Insurance database (SNDS). All PLWHIV present in the database in 2013 were identified.  All healthcare resource consumption from 2008 to 2015 inclusive was documented and costed (for 2013 to 2015) from the perspective of public health insurance. High-cost and high mortality patient profiles were identified by a machine learning algorithm.

RESULTS

In 2013, 96,423 PLWHIV were identified in the SNDS database, including 3,373 incident cases. Overall, 3,224 PLWHIV died during the three-year follow-up period (mean annual mortality rate: 1.1%). The mean annual per capita cost incurred by PLWHIV was € 14,223, corresponding to a total management cost of HIV of € 1,370 million in 2013. The largest contribution came from the cost of antiretroviral medication (M€ 870; 63%) followed by hospitalisation (M€ 154; 11%). The costs incurred in the year preceding death were considerably higher. Four specific patient profiles were identified for under/over-expressing these costs, suggesting ways to reduce them.

CONCLUSIONS

Even though current therapeutic regimens provide excellent virological control in most patients, PLWHIV have excess mortality. Other factors such as comorbidities, lifestyle factors and screening for cancer and cardiovascular disease, need to be targeted in order to lower the mortality and cost associated with HIV infection.

Analyzing the use of artificial intelligence for the management of chronic obstructive pulmonary disease (COPD)

OBJECTIVE

Chronic obstructive pulmonary disease (COPD) is a disease that causes airflow limitation to the lungs and has a high morbidity around the world. The objective of this study was to evaluate how artificial intelligence (AI) is being applied for the management of the disease, analyzing the objectives that are raised, the algorithms that are used and what results they offer.

METHODS

We conducted a scoping review following the Arksey and O'Malley (2005) and Levac et al. (2010) guidelines. Two reviewers independently searched, analyzed and extracted data from papers of five databases: Web of Science, PubMed, Scopus, Cinahl and Cochrane. To be included, the studies had to apply some AI techniques for the management of at least one stage of the COPD clinical process. In the event of any discrepancy between both reviewers, the criterion of a third reviewer prevailed.

RESULTS

380 papers were identified through database searches. After applying the exclusion criteria, 67 papers were included in the study. The studies were of a different nature and pursued a wide range of objectives, highlighting mainly those focused on the identification, classification and prevention of the disease. Neural nets, support vector machines and decision trees were the AI algorithms most commonly used. The mean and median values of all the performance metrics evaluated were between 80% and 90%.

CONCLUSIONS

The results obtained show a growing interest in the development of medical applications that manage the different phases of the COPD clinical process, especially predictive models. According to the performance shown, these models could be a useful complementary tool in the decision-making by health specialists, although more high-quality ML studies are needed to endorse the findings of this study.

A Self-Activated CNN Approach for Multi-Class Chest-Related COVID-19 Detection

Chest diseases can be dangerous and deadly. They include many chest infections such as pneumonia, asthma, edema, and, lately, COVID-19. COVID-19 has many similar symptoms compared to pneumonia, such as breathing hardness and chest burden. However, it is a challenging task to differentiate COVID-19 from other chest diseases. Several related studies proposed a computer-aided COVID-19 detection system for the single-class COVID-19 detection, which may be misleading due to similar symptoms of other chest diseases. This paper proposes a framework for the detection of 15 types of chest diseases, including the COVID-19 disease, via a chest X-ray modality. Two-way classification is performed in proposed Framework. First, a deep learning-based convolutional neural network (CNN) architecture with a soft-max classifier is proposed. Second, transfer learning is applied using fully-connected layer of proposed CNN that extracted deep features. The deep features are fed to the classical Machine Learning (ML) classification methods. However, the proposed framework improves the accuracy for COVID-19 detection and increases the predictability rates for other chest diseases. The experimental results show that the proposed framework, when compared to other state-of-the-art models for diagnosing COVID-19 and other chest diseases, is more robust, and the results are promising.

The Association with COPD Readmission Rate and Access to Medical Institutions in Elderly Patients

Purpose

Up to 20% of patients with chronic obstructive pulmonary disease (COPD) require re-admission within 30 days of discharge after hospitalization for acute exacerbations of the disease. These re-admissions can increase morbidity and the economic burden of COPD. Reducing re-admissions has become a policy target in many developed countries. We investigated the risk factors for COPD re-admissions among older adults with COPD.

Patients and Methods

Data obtained from the National Health Insurance Service-Senior Cohort (NHIS-SC) in Korea were analyzed. The subjects included 558,147 patients aged ≥70 who had been admitted for COPD between 2013 and 2015. Re-admission was defined as being re-hospitalized within 30 days after discharge. The key variables selected from the database included income-based insurance contributions, demographical variables, information on inpatient medical services, types of healthcare facilities, and emergency time relevance index (TRI). The TRI is a regional medical-use analysis index that evaluates whether the capacity of the medical services available is appropriate for the medical needs of the target residents.

Results

In 814 COPD re-admission cases among 4867 total admissions due to COPD in elderly subjects, higher re-admission rates were associated with male sex, admission to district hospitals, medical aid recipients, and a longer hospital stay. When additionally adjusting the TRI to identify the difference in re-admission rates due to medical service accessibility, the same results were found, except for the areas of residence. The TRI was lower in re-admission cases (odds ratio 0.991 [95% CI, 0.984‒0.998], P = 0.013).

Conclusion

In this study, COPD re-admission rates among older adults were significantly associated with sex, length of hospital stay, and the type of hospital. The capacity of the medical services provided was also related to the COPD re-admission rate. Better access to appropriate emergency services is associated with reduction of COPD re-admission rates.

Artificial Intelligence and Machine Learning in Chronic Airway Diseases: Focus on Asthma and Chronic Obstructive Pulmonary Disease

Chronic airway diseases are characterized by airway inflammation, obstruction, and remodeling and show high prevalence, especially in developing countries. Among them, asthma and chronic obstructive pulmonary disease (COPD) show the highest morbidity and socioeconomic burden worldwide. Although there are extensive guidelines for the prevention, early diagnosis, and rational treatment of these lifelong diseases, their value in precision medicine is very limited. Artificial intelligence (AI) and machine learning (ML) techniques have emerged as effective methods for mining and integrating large-scale, heterogeneous medical data for clinical practice, and several AI and ML methods have recently been applied to asthma and COPD. However, very few methods have significantly contributed to clinical practice. Here, we review four aspects of AI and ML implementation in asthma and COPD to summarize existing knowledge and indicate future steps required for the safe and effective application of AI and ML tools by clinicians.

Remote Patient Monitoring for the Detection of COPD Exacerbations

Background

COPD exacerbations occur more frequently with disease progression and are associated with worse prognosis and higher healthcare expenditure.

Purpose

To utilize a networked system, optimized with statistical process control (SPC), for remote patient monitoring (RPM) and to identify potential predictors of COPD exacerbations.

Methods

Seventeen subjects, mean (SD) age of 69.7 (7.2) years, with moderate to severe COPD received RPM. Over 2618 patient-days (7.17 patient-years) of monitoring, we obtained daily symptom scores, treatment adherence, self-reported activity levels, daily spirometry (SVC, FEV₁, FVC, PEF), inspiratory capacity (IC), and oxygenation (SpO₂). These data were used to identify predictors of exacerbations defined using Anthonisen and other criteria.

Results

After implementation of SPC, concordance analysis showed substantial agreement between FVC (decrease below the 7-day rolling average minus 1.645 SD) and self-reported healthcare utilization events (κ=0.747, P<0.001) as well as between increased use of inhaled short-acting bronchodilators and exacerbations defined by two Anthonisen criteria (κ=0.611, P<0.001) or modified Anthonisen criteria (κ=0.622, P<0.001). There was a moderate agreement between FEV₁ (decrease >1.645 SD below the 7-day rolling average) and self-reported healthcare utilization events (κ=0.475, P<0.001) and between SpO₂ less than 90% and exacerbations defined by two Anthonisen criteria (κ=0.474, P<0.001) or modified Anthonisen criteria (κ=0.564, P<0.001).

Conclusion

Exacerbations were best predicted by FVC and FEV₁ below the one-sided 95% confidence interval derived from SPC but also by increased use of inhaled short-acting bronchodilators and fall in oxygen saturation. An RPM program that captures these parameters may be used to guide appropriate interventions aimed at reducing healthcare utilization in COPD patients.