Markov Model Research Articles

Cost-effectiveness analysis of a lung cancer screening program in the Netherlands: a simulation based on NELSON and NLST study outcomes

Background In the Netherlands, lung cancer is the leading cause of cancer-related death, accounting for more than 10,000 annual deaths. Lung cancer screening (LCS) studies using low-dose computed tomography (LDCT) have demonstrated that early detection reduces lung cancer mortality. However, no LCS program has been implemented yet in the Netherlands. A national LCS program has the potential to enhance the health outcomes for lung cancer patients in the Netherlands. Objective and Methods This study evaluates the cost-effectiveness of LCS compared to no-screening in the Netherlands, by simulating the screening outcomes based on data from NEderlands–Leuvens Longkanker Screenings ONderzoek (NELSON) and National Lung Screening Trial (NLST). We simulated annual screening up to 74 years of age, using inclusion criteria from the respective studies. A decision tree and Markov model was used to predict the incremental costs, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratio (ICERs) for the screening population. The analysis used a lifetime horizon and a societal perspective. Results Compared to no-screening, LCS resulted in an ICER of €5,169 per QALY for the NELSON simulation, and an ICER of €17,119 per QALY for the NLST simulation. The screening costs were highly impactful for the cost-effectiveness. The most influential parameter was the CT scan cost. Cost reduction for CT from €201 to €101 per scan would reduce the ICER to €2,335 using NELSON criteria. Additionally, LCS could prevent 15,115 and 12,611 premature lung cancer deaths, accompanied by 1.66 and 1.31 QALYs gained per lung cancer case for the NELSON and NLST simulations, respectively. Conclusion LCS was estimated to be cost-effective in the Netherlands for both simulations at a willingness-to-pay threshold of €20,000 per QALY. Using the NELSON criteria, less than €5,500 per QALY had to be spent. Lowering the cost per CT exam would lead to a further reduction of this amount.

216 Using GPS collars and vaginal temperature sensors for characterizing rangeland usage and climatic stress in sheep

Abstract Increased climatic variability represents the most significant threat to arid rangeland production systems. We set out to describe land use behavior of rangeland sheep and associations between land use phenotypes and body weight (BW) gain via the investigation of GPS data paired with vaginal temperature sensors to characterize fitness under climatic stress. During a fortnight spanning Aug 19th to Sep 1st, 2023, we deployed 200 vaginal temperature sensors and 199 GPS collars on extensively managed Australian Merinos, and a ⅝ Merino X ⅜ Rambouillet composite line in central Nevada, totaling 200 sampled individuals. Vaginal temperature sensors were constructed by affixing Thermochron iButtons to small ruminant CIDRs depleted of progesterone. GPS collars were constructed by enclosing i-gotU GT-120B GPS units in rubber shrink tubing around large dog collars. The iButtons sampled for vaginal temperature every 10 min, and the GPS collars sampled for location every 5 min. Body weights were recorded at the beginning and end of the trial. Remotely sensed data were acquired from the USGS and climateengine.org to analyze covariates such as slope and elevation, and NDVI, and climate, respectively, and were combined with observational data to generate rasters of distance from water and shade. GPS Position was continuously modeled using the R package momentuHMM (function crwWrap) to allow for the union of simultaneous temperature and location observations. A hidden markov model was employed (function fitHMM) to delineate between active and resting observations. Estimates were more significant when modeling land use and temperature phenotypes by just “active” observations. We used a forward progressing model building function with the R package leaps (function regsubsets) to associate weight change with all land use and temperature phenotypes. Logged distance to shade (P = 0.023) and vertical distance traveled (P = 0.057) were associated with BW gain. The square root of the range in vaginal temperature was positively associated with BW gain (P = 0.045) in a simple linear regression model, suggesting that the ability to adapt to heat stress via metabolic elasticity increases fitness. This study also estimated repeatability for land use behavior and temperature traits. The repeatabilities for vertical distance traveled and distance to shade were 0.20 ± 0.03 and 0.04 ± 0.01, respectively. The repeatabilities for mean temperature and vaginal temperature range were 0.47 ± 0.03 and 0.26 ± 0.03, respectively. In conclusion, some land use traits are lowly to moderately repeatable and have significant relationships to BW gain (i.e., fitness) in rangeland settings. Temperature traits are moderately to highly repeatable and can be associated with BW gain.

Cost-Effectiveness Analysis of Routine Childhood Immunization with 20-Valent versus 15-Valent Pneumococcal Conjugate Vaccines in Germany.

This study aimed to evaluate the cost-effectiveness of routine childhood immunization with the 20-valent pneumococcal conjugate vaccine (PCV20) in a four-dose regimen (3 + 1 schedule) versus the 15-valent PCV (PCV15/V114) in a three-dose regimen (2 + 1) in Germany. The study utilized a decision-analytic Markov model to estimate lifetime costs and effectiveness outcomes for a single birth cohort in Germany. The model tracked the incidence of acute pneumococcal infections and long-term pneumococcal meningitis sequelae for both vaccination strategies. The vaccine effectiveness data were derived from published clinical trials and observational studies of PCV7 and PCV13. Indirect effects, such as herd protection and serotype replacement, were included in the model. The model adopted a societal perspective, including direct medical, direct non-medical, and indirect costs. Scenario and sensitivity analyses were performed. In the base case, PCV20 prevented more pneumococcal disease cases and deaths, with an expected gain of 96 quality-adjusted life years (QALYs) compared to V114. However, PCV20 was associated with a total incremental cost of EUR 48,358,424, resulting in an incremental cost-effectiveness ratio (ICER) of EUR 503,620/QALY. Most of the scenario and sensitivity analyses estimated that the ICER for PCV20 exceeded EUR 150,000/QALY. Routine childhood immunization with PCV20 instead of V114 may not be an economically efficient use of healthcare resources in Germany.

Tracking toads from dusk till dawn: linking inter- and intra-individual variation in movement behavior

Abstract Animals can differ in their individual movement behaviors during their daily displacements. Studies of animal movement patterns often disregard the need to understand individual variation in these patterns and the role of this variation in shaping population distributions of movements. To assess the link between individual movements and population distributions, we radio-tracked 13 Fowler’s Toads (Anaxyrus fowleri) in Long Point, Ontario, Canada, after their breeding season. We recorded individual coordinates every 30 min, from the moment they emerged from their burrows around dusk until they burrowed in the morning, and obtained a total of 157 movements. We used the resulting series of step lengths and turning angles in a Hidden Markov Model to estimate movement type as “directed” or “non-directed.” We also assigned to each encounter one of 3 space-dependent movement states: “foraging”, “in transit”, or “searching”, and found that movement type and state switched within individuals irrespective of time during a single night. We find that switching between movement types varied throughout the night, leading to stochastic within-individual variation in movements, and the distribution of movement distances differed significantly among individuals. Movement states, however, were time-dependent, suggesting displacement routines were similar across individuals. Variation in movement behavior within individuals scales up to result in variation among individuals which, in turn, was found to shape the distribution of the sampled population. Our findings therefore underline the importance of a multi-scale approach in the study of movement.

Role of Multistate Models to Predict Patency, Limb Salvage, and Survival: New Concepts to Analyse Data in Peripheral Arterial Disease

In peripheral arterial disease, patency, limb salvage, and survival rates are mostly reported using Kaplan-Meier analyses. When comparing different revascularisation techniques, these methods have limitations in analysing complex patient flows over time. This study aimed to present, illustrate, and discuss new concepts based on multistate models of analysing outcome parameters in peripheral arterial disease. Previously published data from a single centre, randomised controlled trial (RCT) with 218 cases that underwent either vein bypass surgery (bypass group, n = 109) or nitinol stent angioplasty (stent group, n = 109) of long femoropopliteal lesions were re-analysed using non-homogeneous Markov models. A step by step description of the concepts of states, state space, definitions, and illustration of transition probability curves as well as the benefits of multistate models is given. The RCT was registered at ISRCTN.com (ISRCTN18315574). Transition probability curves over time showed similar patterns in the bypass and stent groups. Significant differences in the transition probabilities were found for transitions from primary patency as well as secondary patency to end of patency. The transition probability for patients with preserved primary patency at 24 months who moved to end of patency at 48 months was 19.9% in the stent group vs. 6.4% in the bypass group (p < .001). The proposed method can answer important questions, such as: Did patients after femoropopliteal stenting with preserved primary patency at two years lose their patency more quickly within the following years compared with bypass surgery? and Did stent patients after a re-intervention to maintain patency at one year lose their patency more quickly compared with bypass surgery within the following years? Completely new research questions can now be raised and answered to optimise treatment and follow up strategies; this might lead to better identification of subgroups at higher risk of clinical deterioration following revascularisation procedures.

Cost-effectiveness of operative intervention for mucinous cystic neoplasm: the role of post-operative diabetes.

Small mucinous cystic neoplasms (MCN) of the pancreas are managed with operative resection in otherwise healthy patients; however, postoperative diabetes development is not considered in recommendations for resection. Via probabilistic microsimulation Markov modeling, we assessed clinical and economic implications of laparoscopic distal pancreatectomy (LDP) versus surveillance for non-DM patients with suspected MCN of 2 cm without high-risk or worrisome features. Primary outcomes included quality-adjusted life years (QALYs), medical costs (2021 USD), and incremental cost-effectiveness ratios (ICERs). We conducted sensitivity analyses to evaluate the robustness of our model to changes in input parameters. All analyses were repeated for a population with pre-DM. Surveillance resulted in 6.52 QALYs and $61,200, while LDP accumulated 6.12 QALYs and $63,700. Almost 20% of the LDP cohort developed DM over the first 10 years, compared to 11% of the surveillance cohort. In a pre-DM cohort, LDP remained Dominated in the base case, with over 40% developing DM postoperatively. In sensitivity analyses, surveillance remained the preferred strategy in most iterations for both cohorts. Surveillance for small suspected MCNs without high-risk features is the preferred strategy from a clinical and economic standpoint. Consensus guidelines should consider the long-term implications of postoperative diabetes development following LDP.

Optimizing smart home appliance energy monitoring using Factorial Hidden Markov Models for Internet of Behavior

The energy demand which is increasing globally strongly underlines the necessity of solutions that will improve power efficiency. This study addresses the research problem of enhancing energy management in smart homes through Non-Intrusive Load Monitoring (NILM), a method that enables consumers and companies to optimize their energy usage by disaggregating total household energy consumption into specific appliance-level data. The research focuses on the creation and establishment of an Internet of Behavior (IoB) model that encourages NILM to develop more energy-efficient and clever residential buildings as its main goal. The Factorial Hidden Markov Model (FHMM) is employed as the core methodology in this NILM process, facilitating the disaggregation of overall household energy usage into individual appliance consumption patterns. This method was used over the known Reference Energy Disaggregation Dataset (REDD) for comparing the actual energy consumption of the appliances with the projected estimates. The results are evidence that the FHMM method is an energy disaggregation technique that can provide important information about the customer usage of energy and, hence will help in the strategic shifting of high-energy appliances to non-peak hours for the user. This study's main value lies in the new application of IoB-based NILM methods to increase energy efficiency, which is the real solution for energy costs and the grid peak load problem. The results of the proposed method are compared with other methods and real data. The comparisons show the acceptable response of the proposed method when it is compared with the real data and is better than another method.

Impact of next-generation sequencing vs polymerase chain reaction testing on payer costs and clinical outcomes throughout the treatment journeys of patients with metastatic non-small cell lung cancer.

For patients with metastatic non-small cell lung cancer (mNSCLC), next-generation sequencing (NGS) biomarker testing has been associated with a faster time to appropriate targeted therapy and more comprehensive testing relative to polymerase chain reaction (PCR) testing. However, the impact on payer costs and clinical outcomes during patients' treatment journeys has not been fully characterized. To assess the costs and clinical outcomes of NGS vs PCR biomarker testing among patients with newly diagnosed de novo mNSCLC from a US payers' perspective. A Markov model assessed costs and clinical outcomes of NGS vs PCR testing from the start of testing up to 3 years after. Patients entered the model after receiving biomarker test results and then initiated first-line (1L) targeted or nontargeted therapy (immunotherapy and/or chemotherapy) depending on actionable mutation detection. A few patients with an actionable mutation were not detected by PCR and inappropriately initiated 1L nontargeted therapy. At each 1-month cycle, patients could remain on treatment with 1L, progress to second line or later (2L+), or die. Literature-based inputs included the rates of progression-free survival (PFS) and overall survival (OS), targeted and nontargeted therapy costs, total costs of testing, and medical costs of 1L, 2L+, and death. Per patient average PFS and OS as well as cumulative costs were reported for NGS and PCR testing. In a modeled population of 100 patients (75% commercial and 25% Medicare), 45.9% of NGS and 40.0% of PCR patients tested positive for an actionable mutation. Relative to PCR, NGS was associated with $7,386 in savings per patient (NGS = $326,154; PCR = $333,540) at 1 year, driven by lower costs of testing, including estimated costs of delayed care and nontargeted therapy initiation before receiving test results (NGS = $8,866; PCR = $16,373). Treatment costs were similar (NGS = $305,644; PCR = $305,283). In the PCR cohort, the per patient costs of inappropriate 1L nontargeted therapy owing to undetected mutations were $6,455, $6,566, and $6,569 over the first 1, 2, and 3 years, respectively. Relative to PCR testing, NGS was associated with $4,060 in savings at 2 years and $1,092 at 3 years. Patients who initiated 1L targeted therapy had an additional 5.4, 8.8, and 10.4 months of PFS and an additional 1.4, 3.6, and 5.3 months of OS over the first 1, 2, and 3 years, respectively, relative to those who inappropriately initiated 1L nontargeted therapy. In this Markov model, as early as year 1, and over 3 years following biomarker testing, patients with newly diagnosed de novo mNSCLC undergoing NGS testing are projected to have cost savings and longer PFS and OS relative to those tested with PCR.

Short- and long-term cost-effectiveness of abrocitinib versus dupilumab in adults with moderate-to-severe atopic dermatitis in China

Background Biologics and JAK inhibitors were the most effective innovative systemic treatments for moderate-to-severe atopic dermatitis (AD). However, their cost-effectiveness in China remains unclear. This study aims to compare both the short- and long-term cost-effectiveness of abrocitinib and dupilumab in adults with moderate-to-severe AD from the perspective of the Chinese healthcare system. Methods A hybrid decision tree and Markov model were developed to simulate the costs and health outcomes of interventions on both short-term and long-term horizons. Short- and long-term horizons were employed to reflect the 26-week induction treatment and model the extended 10-year maintenance treatment period, respectively. The cost-effectiveness of strategies was measured by incremental cost-effectiveness ratios (ICERs), which were then compared with the willingness-to-pay threshold (WTP) that was equivalent to the gross domestic product (GDP) per capita of China in 2023 ($12,681 [€11,679.26]). One-way and probabilistic sensitivity analyses were conducted to validate the robustness of the model. Results Over the short-term horizon, the QALYs (quality-adjusted life years) gained were 0.43 for the abrocitinib group and 0.42 for the dupilumab group, with the costs being $2,716.01 (€2,501.46) and $3,940.33 (€3,629.06), respectively. Over the long-time horizon, abrocitinib therapy yields higher QALYs (6.60 versus 6.53) and incurs a lower cost ($22,765.15 [€20,966.81] versus $30,683.38 [€28,259.54]) compared to dupilumab. The probability of abrocitinib being cost-effective was nearly 100% under the current WTP. Both short- and long-term results showed that abrocitinib was more effective and less costly than dupilumab, making abrocitinib the dominant option. Conclusions Abrocitinib was dominant compared to dupilumab both over the short- and long-term horizon for moderate-to-severe AD in China. Future research incorporating real-world evidence and long-term efficacy outcomes could further refine these economic evaluations.

Land/use land /cover dynamics and future scenario of Mayurakshi river basin by random forest and CA–Markov model

Land/use land /cover dynamics and future scenario of Mayurakshi river basin by random forest and CA–Markov model

On-Demand Meal Delivery: A Markov Model for Circulating Couriers

On-demand meal delivery has become a feature of most cities around the world as a result of platforms and apps that facilitate it as well as the pandemic, which for a period, closed restaurants. Meals are delivered by couriers, typically on bikes, e-bikes, or scooters, who circulate collecting meals from kitchens and delivering them to customers, who usually order online. A Markov model for circulating couriers with n + 1 parameters, where [Formula: see text] is the number of kitchens plus customers, is derived by entropy maximization. There is one parameter for each kitchen and customer representing the demand for a courier, and there is one parameter representing the urgency of delivery. It is shown how the mean and variance of delivery time can be calculated once the parameters are known. The Markov model is irreducible. Two procedures are presented for calibrating model parameters on a data set of orders. Both procedures match known order frequencies with fitted visit probabilities; the first inputs an urgency parameter value and outputs mean delivery time, whereas the second inputs mean delivery time and outputs the corresponding urgency parameter value. Model calibration is demonstrated on a publicly available data set of meal orders from Grubhub. Grubhub data are also used to validate the calibrated model using a likelihood ratio. By changing the location of one kitchen, it is shown how the calibrated model can estimate the resulting change in demand for its meals and the corresponding mean delivery time. The Markov model could also be used for the assignment of courier trips to a street network. History: This paper has been accepted for the Transportation Science Special Issue on ISTTT Conference.

A Markov model approach for statistical tropical cyclone wind radii baseline forecasts

Abstract A skill baseline for five-day, 34-, 50-, and 64-knot (1 kt = 0.514 m s−1) tropical cyclone (TC) wind radii forecasts is described. The Markov Model CLiper (MMCL) generates a sequence of 12-h forecasts out to a forecast length limited only by the length of the forecast track and intensity. The model employs a climatology of TC size based on infrared satellite imagery, a Markov chain, and a basin-specific drift. MMCL uses the initial wind radii and initial forecast track and intensity as input. Unlike the previously developed wind radii climatology and persistence model (DRCL) that reverts to a climatological size and shape after approximately 48 h, MMCL retains more of its initial size and asymmetry and is likely more palatable for use in operational forecasting. MMCL runs operationally in the western North Pacific basin, the North Indian Ocean, and the Southern Hemisphere for the Joint Typhoon Warning Center (JTWC) in Pearl Harbor, Hawaii. This work also describes the development of Atlantic and eastern North Pacific versions of MMCL. MMCL’s formulation allows unlimited extension of forecast lead time without reverting to a generic climatological size and shape. Independent forecast comparisons between MMCL and DRCL for the 2020–2022 seasons demonstrates that MMCL’s mean absolute errors are generally smaller and biases are closer to zero in North Atlantic, and eastern North Pacific basins, and in the Southern Hemisphere. This validation includes a few example forecasts and demonstrates that MMCL can be used both as a baseline for assessing wind radii forecast skill and operational use.

Understanding Learning from EEG Data: Combining Machine Learning and Feature Engineering Based on Hidden Markov Models and Mixed Models.

Theta oscillations, ranging from 4-8 Hz, play a significant role in spatial learning and memory functions during navigation tasks. Frontal theta oscillations are thought to play an important role in spatial navigation and memory. Electroencephalography (EEG) datasets are very complex, making any changes in the neural signal related to behaviour difficult to interpret. However, multiple analytical methods are available to examine complex data structures, especially machine learning-based techniques. These methods have shown high classification performance, and their combination with feature engineering enhances their capability. This paper proposes using hidden Markov and linear mixed effects models to extract features from EEG data. Based on the engineered features obtained from frontal theta EEG data during a spatial navigation task in two key trials (first, last) and between two conditions (learner and non-learner), we analysed the performance of six machine learning methods on classifying learner and non-learner participants. We also analysed how different standardisation methods used to pre-process the EEG data contribute to classification performance. We compared the classification performance of each trial with data gathered from the same subjects, including solely coordinate-based features, such as idle time and average speed. We found that more machine learning methods perform better classification using coordinate-based data. However, only deep neural networks achieved an area under the ROC curve higher than 80% using the theta EEG data alone. Our findings suggest that standardising the theta EEG data and using deep neural networks enhances the classification of learner and non-learner subjects in a spatial learning task.

The Serine Acetyltransferase (SAT) Gene Family in Tea Plant (Camellia sinensis): Identification, Classification and Expression Analysis under Salt Stress.

Cysteine plays a pivotal role in the sulfur metabolism network of plants, intimately influencing the conversion rate of organic sulfur and the plant's capacity to withstand abiotic stresses. In tea plants, the serine acetyltransferase (SAT) genes emerge as a crucial regulator of cysteine metabolism, albeit with a notable lack of comprehensive research. Utilizing Hidden Markov Models, we identified seven CssSATs genes within the tea plant genome. The results of the bioinformatics analysis indicate that these genes exhibit an average molecular weight of 33.22 kD and cluster into three distinct groups. Regarding gene structure, CssSAT1 stands out with ten exons, significantly more than its family members. In the promoter regions, cis-acting elements associated with environmental responsiveness and hormone induction predominate, accounting for 34.4% and 53.1%, respectively. Transcriptome data revealed intricate expression dynamics of CssSATs under various stress conditions (e.g., PEG, NaCl, Cold, MeJA) and their tissue-specific expression patterns in tea plants. Notably, qRT-PCR analysis indicated that under salt stress, CssSAT1 and CssSAT3 expression levels markedly increased, whereas CssSAT2 displayed a downregulatory trend. Furthermore, we cloned CssSAT1-CssSAT3 genes and constructed corresponding prokaryotic expression vectors. The resultant recombinant proteins, upon induction, significantly enhanced the NaCl tolerance of Escherichia coli BL21, suggesting the potential application of CssSATs in bolstering plant stress resistance. These findings have enriched our comprehension of the multifaceted roles played by CssSATs genes in stress tolerance mechanisms, laying a theoretical groundwork for future scientific endeavors and research pursuits.

Cost-effectiveness of first-line enfortumab vedotin in addition to pembrolizumab for metastatic urothelial carcinoma in the United States.

The EV-302 trial found that the combination of enfortumab vedotin (EV) with pembrolizumab significantly improved survival for patients with metastatic urothelial carcinoma (mUC). However, given the high cost of the drugs, there is a need to assess its value by considering both efficacy and cost. This study assessed the cost-effectiveness of EV plus pembrolizumab as a first-line treatment for patients with mUC from the perspective of U.S. payers. A Markov model was developed to compare the lifetime costs and effectiveness of EV in combination with pembrolizumab with chemotherapy in the treatment of mUC patients from U.S. payer perspective. Life-years (LYs), quality-adjusted LYs (QALYs), and lifetime costs were estimated. One-way, two-way and probabilistic sensitivity analyses were conducted to evaluate model uncertainty. Additionally, subgroup analyses were performed. Compared to chemotherapy, the combination of EV and pembrolizumab provided an additional 2.10 LYs and 1.72 QALYs, at an incremental cost of $962,240.8 per patient. The incremental cost-effectiveness ratio (ICER) is $558,973 per QALY. Subgroup analysis indicated that patients ineligible for cisplatin treatment had a lower ICER compared to those who were eligible for cisplatin. From the perspective of US payers, at a willingness-to-pay threshold of $150,000 per QALY, the combination of EV and pembrolizumab is estimated to not be cost-effective compared to traditional chemotherapy in the first-line treatment of mUC patients.

Identifying Fishing Trip Behavior from Vessel Monitoring System (VMS) Data Using Machine Learning Models

Illegal fishing in Indonesian waters poses a serious challenge that requires innovative solutions. This research offers an advanced technological approach by applying the Hidden Markov Model (HMM) in Machine Learning to address this issue. Data from the Vessel Monitoring System (VMS) is utilized to efficiently identify fishing vessel activities. By involving a dataset that encompasses various vessel activities, this model can detect suspicious fishing practices in real-time. The research findings demonstrate that this model consistently identifies fishing vessel activities with a high level of accuracy. This study makes a significant contribution to efforts in preventing Illegal, Unreported, and Unregulated (IUU) Fishing and supports marine resource sustainability initiatives.

RESEARCH ON THE CONSTRUCTION OF AI COMPOSITION SYSTEM BASED ON HMM

The Markov chain is an automatic accompaniment algorithm for intelligent computer systems, belonging to the interdisciplinary research field of musicology and computer science. Currently, there are many methods for AI music generation, but research on AI music generation based on the Hidden Markov Model (HMM) is relatively scarce. This paper proposes a method for constructing an AI composition system based on the HMM. This system achieves the goal of automatically generating accompaniment music from score data. The proposed system has achieved relatively stable results in the generation of musical elements such as form and harmony, accompaniment texture, and instrumentation, and has scored well in evaluation experiments.

Separating states in astronomical sources using hidden Markov models: with a case study of flaring and quiescence on EV Lac

ABSTRACT We present a new method to distinguish between different states (e.g. high and low, quiescent and flaring) in astronomical sources with count data. The method models the underlying physical process as latent variables following a continuous-space Markov chain that determines the expected Poisson counts in observed light curves in multiple passbands. For the underlying state process, we consider several autoregressive processes, yielding continuous-space hidden Markov models of varying complexity. Under these models, we can infer the state that the object is in at any given time. The continuous state predictions from these models are then dichotomized with the help of a finite mixture model to produce state classifications. We apply these techniques to X-ray data from the active dMe flare star EV Lac, splitting the data into quiescent and flaring states. We find that a first-order vector autoregressive process efficiently separates flaring from quiescence: flaring occurs over 30 per cent–40 per cent of the observation durations, a well-defined persistent quiescent state can be identified, and the flaring state is characterized by higher plasma temperatures and emission measures.

Motion Classification of Objects using Accelerometer and Gyroscope Readings

It can be challenging to properly analyze data from moving objects when there is limited information about the conditions under which the data was collected. The focus of this research is to use accelerometer and gyroscope data as the input to machine learning models to accurately classify data points based on the movement of the object within its environment. Three machine learning models were implemented and tested: a Hidden Markov Model (HMM), a Random Forest Model (RF), and a Multilayer Perceptron Model (MLP). The majority of the data collected via a simulated environment in Gazebo was used to train all three models. The simulated experiments involved a Jackal Robot from Clearpath Robotics driving on three terrains — flat ground and up and down a ramp with a 7-degree slope — that match the classification states. The simulation data was manually classified based on the delta values between orientation data. The models were tested using the remaining 20% of simulated data, data from a physical Jackal, and data from haul trucks working in an open-pit mine. The HMM is based on unsupervised learning which meant that the pre-classified datapoints were not useful in training the model. This model was ineffective at classifying the datapoints because the probabilities of the classification changing from one datapoint to the next were situationally skewed, so it was highly improbable that the state changed between datapoints. Due to the inability of this model to classify the simulated data, no further testing was completed. The RF model was able to classify the simulated data with 99.78% accuracy and the MLP model classified the simulated data with 99.73% accuracy. The RF model and the MLP model require noisier training data to accurately classify real-world data. Further testing is currently being conducted using data from a physical Jackal and haul trucks.

Chemoradiotherapy plus immunotherapy for locoregionally advanced nasopharyngeal carcinoma: A cost-effectiveness analysis.

Research focused on the addition of immune checkpoint inhibitors (ICIs) to radiotherapeutic regimens in patients with cancer has become increasingly common, revealing promising improvements in efficacy outcomes. In patients with locoregionally advanced nasopharyngeal carcinoma (NPC), combining immunotherapy with chemoradiotherapy can facilitate the significant prolongation of survival, emphasizing the need for pharmacoeconomic studies focused on the clinical uptake of these innovative treatment regimens. A three-state Markov model was developed based on clinical data from the randomized phase 3 CONTINUUM trial and used to compare the cost-effectiveness of chemoradiotherapy plus sintilimab (sintilimab group) to chemoradiotherapy alone (standard group), analyzing outcomes including incremental cost-effectiveness ratio (ICER), incremental net monetary benefit (INMB), and incremental net-health benefit (INHB) values at a willingness-to-pay (WTP) threshold corresponding to three times the Chinese GDP per capita ($37 035 per quality-adjusted life year [QALY]). The total costs for patients in the sintilimab and standard groups (QALYs [LYs]) were $92 116 (6.68 [10.03]) and $53 255 (3.75 [5.55]), respectively, for an ICER of $13 230/QALY ($8672/LY), an INMB of $70 021 with INHB of 1.89 QALYs. Using the selected WTP threshold. On the standard WTP threshold, the prevalence of sintilimab group as the primary treatment was 90.55% in China. The establishment of the model is stable. Adding sintilimab to chemoradiotherapeutic regimens represents an innovative and cost-effective means for patients with locoregionally advanced NPC management in China.