Markov Model Research Articles

MIPPIS: protein–protein interaction site prediction network with multi-information fusion

BackgroundThe prediction of protein–protein interaction sites plays a crucial role in biochemical processes. Investigating the interaction between viruses and receptor proteins through biological techniques aids in understanding disease mechanisms and guides the development of corresponding drugs. While various methods have been proposed in the past, they often suffer from drawbacks such as long processing times, high costs, and low accuracy.ResultsAddressing these challenges, we propose a novel protein–protein interaction site prediction network based on multi-information fusion. In our approach, the initial amino acid features are depicted by the position-specific scoring matrix, hidden Markov model, dictionary of protein secondary structure, and one-hot encoding. Simultaneously, we adopt a multi-channel approach to extract deep-level amino acids features from different perspectives. The graph convolutional network channel effectively extracts spatial structural information. The bidirectional long short-term memory channel treats the amino acid sequence as natural language, capturing the protein’s primary structure information. The ProtT5 protein large language model channel outputs a more comprehensive amino acid embedding representation, providing a robust complement to the two aforementioned channels. Finally, the obtained amino acid features are fed into the prediction layer for the final prediction.ConclusionCompared with six protein structure-based methods and six protein sequence-based methods, our model achieves optimal performance across evaluation metrics, including accuracy, precision, F1, Matthews correlation coefficient, and area under the precision recall curve, which demonstrates the superiority of our model.

Large-scale dependent multiple testing via higher-order hidden Markov models

ABSTRACT Taking into account the local dependence structure in large-scale multiple testing is expected to improve both the efficiency of the testing procedure and the interpretability of scientific findings. The hidden Markov model (HMM), as an effective model to describe the sequential dependence, has been successfully applied to large-scale multiple testing with local correlations. However, in many applications, the first-order Markov chain is not flexible enough to capture the complexity of local correlations. To address this issue, this paper proposes a novel multiple testing procedure that uses a higher-order Markov chain to better characterize local correlations among tests. The proposed procedure is validated by theoretical results and simulation studies, which show that it outperforms its competitors in terms of power. Finally, a real data analysis is presented to demonstrate the favorable performance of the proposed procedure.

Estimated health benefits, costs, and cost-effectiveness of implementing WHO's sodium benchmarks for packaged foods in India: a modelling study.

Excess dietary sodium intake has been associated with death and disability. WHO has released global sodium benchmarks for packaged foods to support countries to reduce population sodium intake. This study aimed to assess the potential health effect, costs, and cost effectiveness of implementing these WHO sodium benchmarks in India. We used a multiple cohort, proportional multistate, life table (Markov) model to estimate the health gains and cost effectiveness for adults if sodium content in packaged foods complied with the WHO benchmarks compared to the status quo. We used India-specific dietary surveys, food composition tables, foods sales data, and sodium content data from packaged food labels to estimate sodium intake before and after the intervention. Data on blood pressure, cardiovascular disease, and chronic kidney disease burden were obtained from the Global Burden of Diseases, Injuries, and Risk Factors study, and the effect of sodium reduction on blood pressure and disease risk was modelled on the basis of meta-analyses of randomised trials and cohort studies. Intervention and health-care costs were used to estimate net costs, and calculate the incremental cost per health-adjusted life-year (HALY) gained. Costs and HALYs were discounted at 3%. In the first 10 years, compliance with the WHO sodium benchmarks was estimated to avert a mean of 0·3 (95% uncertainty interval [UI] 0·2-0·5) million deaths from cardiovascular diseases and chronic kidney disease, a mean of 1·7 (95% UI 1·0-2·4) million incident cardiovascular disease events, and 0·7 (0·4-1·0) million new chronic kidney disease cases, compared with current practice. Over 10 years, the intervention was projected to be cost saving (100·0% probability), generating 1·0 (0·6 to 1·4) billion HALYs and US$0·8 (95% UI 0·3 to 1·4) million in cost savings. Over the population lifetime, the intervention could prevent 4·2 (2·4-6·0) million deaths from cardiovascular diseases and chronic kidney disease, 14·0 (8·2-20·1) million incident cardiovascular disease events, and 4·8 (2·8-6·8) new chronic kidney disease cases, with an 84·2% probability of being cost-saving and 100·0% probability of being cost-effective. Our modelling data suggest a high potential for compliance with WHO sodium benchmarks for packaged food being associated with substantial health gains and cost savings, making a strong case for India to mandate the implementation of the WHO sodium benchmarks, particularly as packaged food consumption continues to rise. WHO Country Office India.

Optimized Architecture for Efficient OFDMA Network Design

This study presents a novel approach to enhancing the design and performance of OFDMA (Orthogonal Frequency Division Multiple Access) networks, with a particular focus on WiMAX (Worldwide Interoperability for Microwave Access) for Best Effort (BE) services. The proposed method integrates a robust Markovian analytical model with four advanced scheduling algorithms: throughput fairness, resource fairness, opportunistic scheduling, and throttling. A sophisticated simulator was developed, incorporating an ON/OFF traffic generator, user-specific wireless channels, and a dynamic central scheduler to validate the model’s accuracy and evaluate its robustness by dynamically allocating radio resources per frame. The validation study showed that the proposed model reduced simulation time by over 90%, completing analytical calculations in just 15 min, compared to nearly 2 days for simulations using conventional scheduling algorithms. Performance metrics such as the average number of active users and resource utilization closely matched those from the validation study, confirming the model’s accuracy. In the robustness study, the model consistently performed well across diverse traffic distributions (exponential and Pareto) and channel conditions. The proposed architecture increased network throughput by up to 25% and reduced latency under dynamic conditions, demonstrating its scalability, adaptability, and efficiency as a crucial solution for next-generation wireless communication systems.

Potential climate predictability of renewable energy supply and demand for Texas given the ENSO hidden state.

Climate variability influences renewable electricity supply and demand and hence system reliability. Using the hidden states of the sea surface temperature of tropical Pacific Ocean that reflect El Niño-Southern Oscillation (ENSO) dynamics that is objectively identified by a nonhomogeneous hidden Markov model, we provide a first example of the potential predictability of monthly wind and solar energy and heating and cooling energy demand for 1 to 6 months ahead for Texas, United States, a region that has a high penetration of renewable electricity and is susceptible to disruption by climate-driven supply-demand imbalances. We find a statistically significant potential for oversupply or undersupply of energy and anomalous heating/cooling demand depending on the ENSO state and the calendar month. Implications for financial securitization and the potential application of forecasts are discussed.

Exploring the Spatially Compounding Multi‐Sectoral Drought Vulnerabilities in Colorado's West Slope River Basins

AbstractThe state of Colorado's West Slope Basins are critical headwaters of the Colorado River and play a vital role in supporting Colorado's local economy and natural environment. However, balancing the multi‐sectoral water demands in the West Slope Basins while maintaining crucial downstream deliveries to Lake Powell is an increasing challenge for water managers. Internal variability of the hydroclimatic system and climate change complicate future vulnerability assessments. This work contributes a detailed accounting of multi‐sectoral drought vulnerability in the West Slope Basins and the impacts of drought on downstream deliveries. We first introduce a novel multi‐site Hidden Markov Model (HMM)‐based synthetic streamflow generator to create an ensemble of streamflows for all West Slope basins that better characterizes the region's drought extremes. We capture the effects of climate change by perturbing the HMM to generate an ensemble of streamflows reflecting plausible changes in climate. We then route both ensembles through StateMod, Colorado's water allocation model, to evaluate spatially compounding drought impacts across the West Slope Basins. Our results illustrate how drought events emerging from the system's stationary internal variability in the absence of climate change can significantly impact local water uses and deliveries to Lake Powell, exceeding extreme conditions in the historical record. Further, we find that even modest climate change can cause a regime shift where historically low downstream delivery volumes and extreme drought impacts become routine. These results can inform future Colorado River planning efforts, and our methodology can be expanded to other snow‐dominated regions that face persistent droughts.

UTILIZING DISCRETE HIDDEN MARKOV MODELS TO ANALYZE TETRAPLOID PLANT BREEDING

In plant heredity, the phenotype is the result of observation that can be directly observed, while the genotype is the underlying hidden factor that underlies the expression of the phenotype. The genotype is an important aspect that needs to be understood to explain the pattern of trait inheritance and predict trait inheritance in subsequent generations. The discrete hidden Markov model is a model generated by pair of an unobserved Markov chain and an observation process. This model can be applied to tetraploid plant crosses by modeling genotypes as hidden state and phenotypes as the obeservation process. The probability of dominant phenotype in monohybrid, dihybrid and trihybrid crosses occurring over ten generations during that period is as follows 61,305%, 37,583%, and 23,041%. Furthermore, as more traits are crossed, the probability of dominant phenotype appearing within ten generations decreases. When the dominant phenotype occurs over ten generations, the same genotype can be obtained in monohybrid, dihybrid, and trihybrid crosses, which is heterozygous in the first and second generations, while from the third to the tenth generation it is homozygous dominant.

Cost-effectiveness of hearing and vision support for residents with dementia in long-term care in Ireland: A Health Economic Analysis Plan (HEAP) for a simulation model for the SENSE-Cog Care feasibility trial

Introduction Sensory challenges exacerbate the dementia symptoms of nursing home residents. The Sense-Cog Care trial is piloting a multi-faceted intervention to investigate whether optimising and supporting hearing, vision, and sensory-friendly environments through the “sensory champion” model in Irish nursing homes can improve quality of life and dementia-related outcomes and be cost-effective (Connelly et al., 2023). A Health Economic Analysis Plan (HEAP) provides a standardised and comprehensive framework by outlining the systematic approach, methodologies, and key considerations involved in assessing the cost-effectiveness of a healthcare intervention. The HEAP presented in this paper describes the alongside-trial decision modelling approach used for preliminary estimates of cost-effectiveness and to inform the design of the final intervention. Methods A Markov model will be used to simulate residents' health progression with and without the intervention. Costs will be estimated from the Irish health and social care perspective. A key output of the planned economic analysis will be the Incremental-Cost-Effectiveness Ratio (ICER) which describes the cost over one additional unit of Quality-Adjusted Life Year (QALY). A Probabilistic Sensitivity Analysis (PSA) will be conducted in the form of a Monte Carlo simulation to examine parameter uncertainty and the probability of a sensory champion being cost-effective. Results Tables for inputs and outputs are presented, and graphical representations for uncertainty, such as a tornado plot and a Cost-Effectiveness Analysis Curve (CEAC), will be produced. The results will be interpreted in the context of the Irish cost-effectiveness thresholds of €20,000 and €45,000 per QALY. Discussion To our knowledge, this is the first study to model the cost-effectiveness of a sensory intervention for nursing home residents with dementia. We aim to develop an early-stage Bayesian model that serves as a prior for further research.

Parkinson’s Disease Interstage Prognostic Biomarkers for Early Detection through Hybrid Machine Learning Model

Background: Parkinson’s Disease (PD) is a neurodegenerative condition that takes 15 to 20 years to exhibit the symptoms. However, during the initial stages of the disease, there are microlevel changes in the body that are typically undetected. The pathogenesis of Parkinson’s disease requires understanding of these microlevel changes. Objectives: The objective of the present study is to develop a hybrid ensembled machine learning pipeline model for identifying inter-stage Parkinson’s Disease (PD) biomarkers for understanding disease progression as well as etiology. Methods: The proposed work was carried out on the dataset GSE202667 from Gene Expression Omnibus, containing time-resolved RNA signatures of CD4+ T cells at various stages. Differentiating genes were identified in different interstage groups. Two types of unsupervised learning methods- distance-based (K-means, Agglomerative, Density-Based Spatial Clustering of Application with Noise) and probability-based (Hidden Markov Model, Gaussian Mixture Model Latent Dirichlet Allocation) were applied. The best algorithms were selected and applied to optimize clusters. Enrichment analysis was conducted on the top 10 PD biomarkers in each category. Findings: The top 10 PD biomarkers in each category are identified and gene set enrichment analysis resulted into their enrichment in three KEGG Pathways and depletion in two GO molecular functions. These biomarkers’ depletion is observed in 215 Reactome pathways and enrichment in 18 Reactome pathways. Twelve GO-Cellular components had an enrichment whereas 111 GO-Cellular components had a depletion in the gene set. A total of 25 GO-Biological process components were enriched, while 339 GO-Biological process components were depleted. Novelty: The proposed hybrid ensembled machine learning pipeline model works as a tool to identify Parkinson’s Disease biomarkers from omics data. The model contributes to identify implicit patterns in omics data in order to unveil Parkinson’s disease progress mechanism through biomarkers discovery. Keywords: Gaussian Mixture Model clustering, Early Detection, k-means clustering, Machine learning, Parkinson’s Disease Interstage Biomarkers

Assessing the Economic Feasibility of Needle-Retractable Safety Syringes (NRSS) Adoption in Indian Healthcare Facilities

Description: Globally, sixteen billion injections are given each year, 95.00% of which are for medicinal purposes. This global injection burden is accounted for by India between 24.5% and 31.0%, and alarmingly, over 62.50% of these injections are performed improperly or are repeated. Objectives: The purpose of this study is to assess the increased costs incurred by transitioning from traditional disposable syringes to Needle-Retractable Safety Syringes (NRSS) for beneficial usage in India. Methods: Utilizing a decision tree model, we evaluated the occurrence of needle-stick injuries and syringe reuse among healthcare personnel and patients. We calculated the lifetime costs associated with HIV - Human Immunodeficiency Virus, HBV - Hepatitis B Virus, and HCV - Hepatitis C Virus infections using three Markov models. Results: NRSS offer substantial benefits by reducing needle-stick injuries and preventing blood-borne infections. Among the options analyzed, Reuse Prevention (RUP) syringes emerged as the most cost-effective choice. In contrast, Single-Use Prevention (SIP) and NRSS syringes are currently not deemed cost-effective at their prevailing prices. We recommend prioritizing RUP syringes and exploring strategies to make them more affordable for broader accessibility. Conclusion: Our analysis suggests that, in the Indian context, RUP syringes present an economically viable solution. However, SIP and NRSS syringes do not currently offer cost-effectiveness. To enhance the cost-effectiveness of NRSS, concerted efforts should be directed towards reducing their price. Abbreviations: NRSS, Needle-Retractable Safety Syringes; RUP, Reuse Prevention; SIP, Single-Use Prevention; NSI, Needle Stick Injuries; BBI, Blood-borne Infections; QOL, Quality-of-life; BCC, Behavior Change Communication.

An evaluation of vilobelimab (anti-C5a) as a cost-effective option to treat severely ill mechanically ventilated patients with COVID-19.

In an effort to expedite the publication of articles, AJHP is posting manuscripts online as soon as possible after acceptance. Accepted manuscripts have been peer-reviewed and copyedited, but are posted online before technical formatting and author proofing. These manuscripts are not the final version of record and will be replaced with the final article (formatted per AJHP style and proofed by the authors) at a later time. COVID-19 patients in intensive care units (ICUs) requiring invasive mechanical ventilation (IMV) have few available treatment options. PANAMO, a multicenter, double-blind, randomized, placebo-controlled phase 3 study of vilobelimab, which blocks the inflammatory process caused by complement component 5a, demonstrated a significant mortality benefit at 28 and 60 days in these patients. A cost-effectiveness analysis was conducted to assess the incremental cost per quality-adjusted life-year (QALY). A Markov model was used to estimate QALYs and the incremental cost-effectiveness ratio (ICER) of vilobelimab plus standard of care (SOC) versus SOC alone. The model simulated progression from severe COVID-19 to survival or death over a lifetime horizon. Outcomes data (COVID-19 all-cause mortality and renal replacement therapy) were incorporated from the PANAMO trial. COVID-19 mortality estimates were based on Centers for Disease Control and Prevention age-specific survival data. Utility values and hospital costs came from the literature. Vilobelimab cost was obtained from RED BOOK Online. For COVID-19 ICU patients, total costs of care were $103,414 (SOC) and $132,247 (SOC plus vilobelimab), respectively, resulting in an incremental cost of $28,833. SOC provided 6.70 QALYs versus 7.99 QALYs for vilobelimab, an additional 1.29 QALYs. The ICER for vilobelimab plus SOC versus SOC alone was $22,287/QALY. Probabilistic sensitivity analysis demonstrated the robustness of the cost-effectiveness result as vilobelimab plus SOC was favored at a willingness-to-pay threshold of $50,000 in over 81% of iterations. Vilobelimab provides a cost-effective option to treat ICU patients with severe COVID-19 receiving IMV compared to SOC, at well below the commonly accepted $50,000 US willingness-to-pay threshold.

Cost-effectiveness analysis of emergency department-based hepatitis C screening and linkage-to-care program

BackgroundIn the United States (US), hepatitis C virus (HCV) screening is not covered by payers in settings outside of primary care. A non-traditional, emergency department (ED)-based HCV screening program can be cost-effective and identify infection in vulnerable populations with a high HCV risk. This study examined the long-term cost-effectiveness of routine HCV screening and linkage-to-care for high-risk patients in the ED from the payer’s perspective.MethodsThe University of Illinois Hospital and Health Sciences System (UIH) implemented Project HEAL (HIV & HCV Screening, Education, Awareness, Linkage-to-Care). Under this initiative, patients who presented to the ED received opt-out HCV screening if they were at high risk for HCV infection (birth cohort between 1945 and 1964, persons who inject drugs, and HIV infection) with subsequent linkage-to-care if infected. Using the summary data from Project HEAL, a hybrid decision-analytic Markov model was developed based on the HCV screening procedure in the ED and the natural history of HCV. A 30-year time horizon and 1-year cycle length were used. All patients who received the ED-based HCV screening were referred for treatment with direct-acting antiviral (DAA) regardless of their fibrosis stage.ResultsWhen unscreened/untreated patients received DAA treatment at F1, F2, F3, and compensated cirrhosis stages, the incremental cost-effectiveness ratio (ICER) ranged from $6,084 to $77,063 per quality-adjusted life year (QALY) gained. When unscreened/untreated patients received DAA treatment at the decompensated cirrhosis stage, no HCV screening was dominated.ConclusionED-based HCV screening and linkage-to-care was cost-effective at the willingness-to-pay (WTP) threshold of $100,000/QALY in all scenarios. A reduction in infected persons in the community may provide additional benefits not evaluated in this study.

GPS-Based Hidden Markov Models to Document Pastoral Mobility in the Sahel

In agrarian systems where animal mobility is crucial for feed management, nutrient cycles and household economy, there is a notable lack of precise data on livestock mobility and herding practices. We introduce a methodology leveraging GPS-based behavioural models to analyse and document pastoral mobility in the Sahel. Over 2.5 years, we conducted a continuous collection of GPS data from transhumant and resident cattle herds in the Senegalese agropastoral semiarid rangelands. We developed a Hidden Markov Model robustly fitted to these data to classify recordings into three states of activity: resting (47% overall), foraging (37%) and travelling (16%). We detail our process for selecting the states and testing data subsets to guide future similar endeavours. The model describes state changes and how temperature affects them. By combining the resulting dataset with satellite-based land-use data, we show the distribution of activities across landscapes and seasons and within a day. We accurately reproduced key aspects of cattle mobility and characterised rarely documented features of Sahel agropastoral practices, such as transhumance phases, nocturnal grazing and in-field rainy season paddocking. These results suggest that our methodology, which we make available, could be valuable in addressing issues related to the future of Sahelian pastoralism.

A cost-effectiveness analysis of three surgical options for treating displaced femoral neck fractures in active older patients in Japan: A full economic evaluation.

For older patients with displaced femoral neck fractures, in which primary osteosynthesis is usually not indicated, there are three primary prosthetic options-bipolar hemiarthroplasty (BHA), single-bearing total hip arthroplasty (SB-THA), and dual-mobility THA (DM-THA). However, the optimal choice for managing displaced femoral neck fractures remains controversial. Accordingly, this study aimed to evaluate the cost-effectiveness of BHA, SB-THA, and DM-THA in active older patients with displaced femoral neck fractures. A decision tree combined with a Markov model was employed to analyze the cost and quality-adjusted life years (QALYs) of BHA, SB-THA, and DM-THA for the management in the Japanese healthcare system. By simulating the five-year trajectory of a 75-year-old woman treated for a displaced femoral neck fracture, the cost-effectiveness of the three surgical options was evaluated. One-way sensitivity analysis and probabilistic sensitivity analysis (PSA) were used to assess parameter uncertainty. Additionally, two scenario analyses were conducted for other settings. The treatment was considered to be cost-effective when the incremental cost-effectiveness ratio (ICER) was below the 5,000,000 yen/QALY threshold. Compared with BHA, SB-THA exhibited higher costs but greater health benefits, resulting in an ICER of 1,499,440 yen/QALY. DM-THA offered additional health benefits compared with SB-THA, with an ICER of 4,145,777 yen/QALY. One-way sensitivity analysis revealed some influential parameters. PSA indicated that the probability of DM-THA, SB-THA, and BHA being cost-effective was 40.1%, 38.5%, and 21.4%, respectively. SB-THA was more cost-effective than BHA in patients aged 65-85 years, while DM-THA was more cost-effective than SB-THA in patients aged 65-75 years. The results suggest that SB-THA is a cost-effective alternative to BHA for displaced femoral neck fractures in active older patients, whereas DM-THA is more cost-effective than SB-THA in relatively younger patients. It is, therefore, recommended that orthopedic surgeons select the most appropriate surgical option based on the individual patient's physiological age.

Cost-Effectiveness of Bivalent Respiratory Syncytial Virus Prefusion F Vaccine for Prevention of Respiratory Syncytial Virus Among Older Adults in Greece

Background/Objectives: To evaluate the health benefits, costs, and cost-effectiveness of vaccination with bivalent respiratory syncytial virus stabilized prefusion F vaccine (RSVpreF) for the prevention of lower respiratory tract disease caused by respiratory syncytial virus (RSV) in Greek adults 60 years of age and older. Methods: A Markov model was adapted to simulate lifetime risk of health and economic outcomes from the public payer’s perspective over a lifetime horizon. Epidemiology, vaccine effectiveness, utilities, and direct medical costs (EUR, 2024) were obtained from published studies, official sources, and local experts. Model outcomes included the number of medically attended RSV cases, stratified by care setting (i.e., hospital, emergency department [ED], outpatient visits [OV]), and attributable RSV-related deaths, costs, life years (LY), quality-adjusted life-years (QALY), and incremental cost-effectiveness ratios (ICERs) of RSVpreF vaccination compared with no vaccination. Results: The model projected 258,170 hospitalizations, 112,248 ED encounters, 1,201,604 OV, and 25,463 deaths related to RSV in Greek older adults resulting in direct medical costs of EUR 1.6 billion over the lifetime horizon. Assuming RSV vaccination would reach the same coverage rates as pneumococcal and influenza programmes, 18,118 hospitalizations, 7874 ED encounters, 48,079 OV, and 1706 deaths could be prevented over the modelled time horizon. The health benefits associated with RSVpreF contributed to an incremental gain of 10,976 LYs and 7230 QALYs compared with no vaccination. The incremental analysis reported that vaccination with RSVpreF was estimated to be a cost-effective strategy resulting in ICERs of EUR 12,991 per LY gained, EUR 19,723 per QALY gained, and EUR 7870 per hospitalized RSV case avoided compared with no vaccination. Conclusions: Vaccination with RSVpreF was a cost-effective strategy for the prevention of RSV disease in Greek adults over 60 years of age. The introduction of RSV vaccination can improve public health by averting RSV cases and deaths and has the potential to fulfil an unmet medical need.

Longitudinal latent overall toxicity (LOTox) profiles in osteosarcoma: a new taxonomy based on latent Markov models

AbstractDue to the presence of multiple types of adverse events (AEs) with different levels of severity, the analysis of longitudinal toxicity data is a difficult task in cancer research. The current literature primarily relies on descriptive-based methods and lacks models that can effectively quantify the overall toxic burden experienced by patients over treatment without losing details of the impact of each AE. In this work, a novel taxonomy based on latent Markov models and compositional data techniques is proposed to model the Latent Overall Toxicity (LOTox) condition of each patient over cycles of treatment. Starting from observed categories of severity of multiple toxicities, the goal is to delineate distinct LOTox conditions and retrieve patients’ probabilities of being in a specific condition at a given cycle, as well as their risk of experiencing “worse" overall toxicity statuses compared to a reference “good" toxic condition. The proposed approach is applied to longitudinal toxicity data from the MRC BO06/EORTC 80931 randomized controlled trial for patients with osteosarcoma. The population of interest includes 377 patients who had successfully completed the six-cycle treatment. Personal characteristics and observed information on six toxicities are used to infer the unobserved LOTox status over the six cycles of chemotherapy. Provided that longitudinal toxicity data are available, the developed procedure is a flexible approach that can be adapted and applied to other cancer studies.

Forecast Analysis of the Stock Market Based on Hidden Markov Model and Long Short-Term Memory Model- Taking the S and P500 Index as an Example

This paper proposes a hybrid method that combines the advantages of HMM and LSTM models to improve the accuracy of stock market predictions, using the S&P500 index as a case study. HMMs are used to identify and model potential market states, while LSTMs predict future stock prices based on these states. Through this integration, this paper hopes to take advantage of HMM’s ability to model market conditions and LSTM’s advantages in forecasting to provide a more robust forecasting framework. This article is mainly divided into five parts. The first part is an introduction to the research background and objectives. The second part is the review and arrangement of relevant literature. The third part is the elaboration of the theoretical basis and the collection and analysis of data. The fourth part is an explanation of the data analysis results. The fifth part is the conclusion and suggestions of this study.

Single-Channel Sleep EEG Classification Method Based on LSTM and Hidden Markov Model

Background: The single-channel sleep EEG has the advantages of convenient collection, high-cost performance, and easy daily use, and it has been widely used in the classification of sleep stages. Methods: This paper proposes a single-channel sleep EEG classification method based on long short-term memory and a hidden Markov model (LSTM-HMM). First, the single-channel EEG is decomposed using wavelet transform (WT), and multi-domain features are extracted from the component signals to characterize the EEG characteristics fully. Considering the temporal nature of sleep stage changes, this paper uses a multi-step time series as the input for the model. After that, the multi-step time series features are input into the LSTM. Finally, the HMM improves the classification results, and the final prediction results are obtained. Results: A complete experiment was conducted on the Sleep-EDFx dataset. The results show that the proposed method can extract deep information from EEG and make full use of the sleep stage transition rule. The proposed method shows the best performance in single-channel sleep EEG classification; the accuracy, macro average F1 score, and kappa are 82.71%, 0.75, and 0.76, respectively. Conclusions: The proposed method can realize single-channel sleep EEG classification and provide a reference for other EEG classifications.

Economic evaluation of Polygenic Risk Scores in healthcare interventions: a systematic review

Abstract Background Polygenic Risk Scores (PRS) quantify genetic susceptibility to diseases, promising personalized healthcare. This review explores current evidence regarding the economic evaluation of strategies based on PRS or other polygenic risk stratification approaches, scrutinizing their methodologies. Methods The study protocol was registered in PROSPERO (CRD42023442780). A systematic search in PubMed, Scopus, and Web of Science identified full economic evaluations of intervention based on polygenic risk stratification strategies. The quality of the included articles was assessed using the Drummond checklist. This study was supported by the EC and MUR under PNRR - M4C2-I1.3 Project PE_00000019 ‘HEAL ITALIA”. Results Nineteen articles were included in the analysis, with oncological conditions being the most frequently investigated (13), followed by cardiovascular conditions (3). In nearly 80% of the studies, PRS was employed for screening interventions, with the general population being the primary target in 14 out of the 19 studies. All the economic analysis models investigated cost-utility in terms of QALYs, with Markov models and microsimulations being the most common structures. The majority of these models were based on simulated cohorts derived mostly from North American or European data, with 9 adopting a healthcare system perspective and 6 a societal perspective. Although delivery strategies for PRS testing were rarely addressed, PRS costs were included in nearly all studies. Indirect costs were examined in fewer than half of the studies. In 12 out of 19 studies, the conclusions claimed the cost utility of PRS involving strategies. Conclusions Despite the technique’s potential, evaluations of various strategies based on PRS approaches yield heterogeneous conclusions regarding cost-utility. Our study highlights the factors contributing to this heterogeneity and underscores the need for further exploration, ideally prioritizing real-world data. Key messages • PRS strategies evaluations vary in cost-utility conclusions, highlighting the need for further investigation. • Real-world data and practical implications should be prioritised in PRS research.

Multiple myeloma: What is the most cost-effective imaging strategy for initial detection of bone lesions?

To determine the cost-effectiveness of different imaging modalities for initial detection of multiple myeloma (MM)-defining bone lesions. A Markov model from the health care system perspective for patients with MGUS was used to evaluate the incremental cost-effectiveness of five imaging techniques: skeletal survey (SS), low-dose computed tomography (LDCT), positron emission computed tomography (PETCT), and whole-body magnetic resonance imaging (WBMRI) with and without diffusion (DIFF). Model inputs, including probabilities, utilities, and costs were obtained from comprehensive literature review. Costs were estimated in 2024 U.S. dollars, effectiveness was measured in quality adjusted life years (QALYs), willingness-to-pay (WTP) threshold was set to $100,000/QALY, and timeframe of the simulation was 20years. Model analyses included Monte Carlo microsimulation and probabilistic sensitivity analysis (PSA). The most cost-effective imaging strategy was dependent on the number of patient risk factors for progression from MGUS to myeloma. At a WTP threshold of $100,000, for patients with no risk factors for progression, LDCT amassed the greatest net monetary benefit (NMB) ($1,030,913.57) while incurring the second lowest costs ($44,870.73). For patients with 1 or 2 risk factors for progression, WBMRI + DIFF amassed the greatest NMB (1 risk factor: $802,637.30, 2 risk factors: $664,430.36). WBMRI and PETCT were absolutely dominated in all cases. PSA also found that the most cost-effective strategy was dependent on the WTP threshold. Our model suggests that LDCT and WBMRI + DIFF can be the most cost-effective imaging strategies for the initial diagnosis of MM in patients, depending on the number of risk factors for progression.