Unstable Clusters Research Articles

A machine-learning regional clustering approach to understand ventilator-induced lung injury: a proof-of-concept experimental study

BackgroundThe spatiotemporal progression and patterns of tissue deformation in ventilator-induced lung injury (VILI) remain understudied. Our aim was to identify lung clusters based on their regional mechanical behavior over space and time in lungs subjected to VILI using machine-learning techniques.ResultsTen anesthetized pigs (27 ± 2 kg) were studied. Eight subjects were analyzed. End-inspiratory and end-expiratory lung computed tomography scans were performed at the beginning and after 12 h of one-hit VILI model. Regional image-based biomechanical analysis was used to determine end-expiratory aeration, tidal recruitment, and volumetric strain for both early and late stages. Clustering analysis was performed using principal component analysis and K-Means algorithms. We identified three different clusters of lung tissue: Stable, Recruitable Unstable, and Non-Recruitable Unstable. End-expiratory aeration, tidal recruitment, and volumetric strain were significantly different between clusters at early stage. At late stage, we found a step loss of end-expiratory aeration among clusters, lowest in Stable, followed by Unstable Recruitable, and highest in the Unstable Non-Recruitable cluster. Volumetric strain remaining unchanged in the Stable cluster, with slight increases in the Recruitable cluster, and strong reduction in the Unstable Non-Recruitable cluster.ConclusionsVILI is a regional and dynamic phenomenon. Using unbiased machine-learning techniques we can identify the coexistence of three functional lung tissue compartments with different spatiotemporal regional biomechanical behavior.

Prenucleation Cluster Pathway is Inconsistent with CaCO3 Kinetics.

It has been debated whether CaCO3 nucleates classically with the attainment of a critical cluster size or nonclassically with the restructuring of a prenucleation cluster (PNC). Here, we determine from the nucleation kinetics of CaCO3 that the transition state is composed of about 10 formula units, irrespective of the supersaturation. Crucially, the size of the transition state is considerably smaller than the average PNC size estimated from experimental characterization. This size discrepancy suggests the PNCs are uninvolved in nucleation, and the kinetics indicate that if CaCO3 nucleates classically, the transition state must be an abnormally unstable (antimagic) cluster.

CCD PHOTOMETRY OF TRAPEZIA STARS I

We present photometric CCD observations of stars in four stellar trapezia ADS 15184, ADS 4728, ADS 2843, and ADS 16795. This study is performed on images obtained at the Observatorio Astronómico Nacional at San Pedro Mártir (OAN), Baja California, México. In this work we utilise aperture photometry to measure the U, B, V, R and I magnitudes of some of the stars in these dynamically unstable stellar clusters (trapezia). Using the Q = (U - B) - 0.72(B - V) parameter we obtained the spectral type of the studied stars as well as their distance to the Sun and their reddening. Slight differences between the Q-derived Spectral types and those listed in SIMBAD might be due to a different value, from 0.72, for the slope of the reddening line on the two-colour diagram.

Cavity-Directed Synthesis of Labile Polyoxometalates for Catalysis in Confined Spaces.

The artificial microenvironments inside coordination cages have gained significant attention for performing enzyme-like catalytic reactions by facilitating the formation of labile and complex molecules through a "ship-in-a-bottle" approach. Despite many fascinating examples, this approach remains scarcely explored in the context of synthesizing metallic clusters such as polyoxometalates (POMs). The development of innovative approaches to control and influence the speciation of POMs in aqueous solutions would greatly advance their applicability and could ultimately lead to the formation of elusive clusters that cannot be synthesized by using traditional methods. In this study, we employ host-guest stabilization within a coordination cage to enable a novel cavity-directed synthesis of labile POMs in aqueous solutions under mild conditions. The elusive Lindqvist [M6O19]2- (M=Mo or W) POMs were successfully synthesized at room temperature via the condensation of molybdate or tungstate building blocks within the confined cavity of a robust and water-soluble Pt6L4(NO3)12 coordination cage. Importantly, the encapsulation of these POMs enhances their stability in water, rendering them efficient catalysts for environmentally friendly and selective sulfoxidation reactions using H2O2 as a green oxidant in a pure aqueous medium. The approach developed in this paper offers a means to synthesize and stabilize the otherwise unstable metal-oxo clusters in water, which can broaden the scope of their applications.

Synergistic bimetallic nanozymes of Ni/ZIF-8 and Cu/ZIF-8 as carbonic anhydrase mimics

The remarkable diversity of metal nodes and ligands, coupled with the extensive array of connection modes, enables the precise design and assembly of metal-organic frameworks (MOFs), which have garnered significant attention in the field of nanozyme. In this study, the synthesis of zeolite imidazole framework materials (ZIFs) with varying Zn2+ and Ni2+ or Cu2+ concentrations were investigated and regulated by methanol, aiming to mimic the CO2 hydration activity of carbonic anhydrase (CA). The materials were comprehensively characterized using TEM, XRD, FTIR, UV-Vis DRS, XPS, and subsequently the catalytic performance of the materials was assessed. The results demonstrated that the structural characteristics of bimetallic Ni/ZIF-8 and Cu/ZIF-8 were influenced by varying concentrations of metal ions. Notably, ZIF-8 doped with lower metal content exhibited enhanced stability, while a precursor containing nickel and copper ions exceeding 50% hindered the formation of the desired ZIFs, resulting in an unstable cluster structure. However, methanol can counteract the inhibitory effect of metal ions on the structure. The bimetallic synergies of doped metals in ZIFs CA mimics can simultaneously expedite both two rate-limiting steps in the CO2 hydration process. The esterase activity of ZIF-8 doped with a small quantity of Ni2+ or Cu2+ can be achieved at a level comparable to that of human carbonic anhydrase (hCA II) at 80 °C, along with remarkable catalytic performance and hydrothermal stability. Specifically, Cu66%/ZIF-8 exhibited an activity of 0.31 U·mg−1 at 25 °C, which was 2.31 times higher than that of ZIF-8, moreover, the esterase activity at 80 °C was approximately 2.21 times greater than that of ZIF-8 under the same condition. While the relative activity of Ni50%/ZIF-8 and Cu50%/ZIF-8 still exceeding 50% even after five repeated utilization cycles. The bimetallic Ni/ZIF-8 and Cu/ZIF-8 can be easily synthesized using a simple method, resulting in enhanced catalytic activity. Therefore, it holds great potential for future applications in CO2 capture.

Weakly nonlinear analysis of thermoacoustic oscillations in can-annular combustors

Can-annular combustors feature clusters of thermoacoustic eigenvalues, which originate from the weak acoustic coupling between $N$ identical cans at the downstream end. When instabilities occur, one needs to consider the nonlinear interaction between all $N$ modes in the unstable cluster in order to predict the steady-state behaviour. A nonlinear reduced-order model for the analysis of this phenomenon is developed, based on the balance equations for acoustic mass, momentum and energy. Its linearisation yields explicit expressions for the $N$ complex-valued eigenfrequencies that form a cluster. To treat the nonlinear equations semianalytically, a Galerkin projection is performed, resulting in a nonlinear system of $N$ coupled oscillators. Each oscillator represents the dynamics of a global mode that oscillates in the whole can-annular combustor. The analytical expressions of the equations reveal how the geometrical and thermofluid parameters affect the thermoacoustic response of the system. To gain further insights, the method of averaging is applied to obtain equations for the slow-time dynamics of the amplitude and phase of each mode. The averaged system, whose solutions compare very well with those of the full oscillator equations, is shown to be able to predict complex transient dynamics. A variety of dynamical states are identified in the steady-state oscillatory regime, including push–push (in-phase) and spinning oscillations. Notably, the averaged equations are able to predict the existence of synchronised states. These states occur when the frequencies of two (or more) unstable modes with nominally different frequencies lock onto a common frequency as a result of nonlinear interactions.

Transient stability enhancement control strategy based on grid dispatching for multi-paralleled grid forming inverters during LVRT

In this article, the transient synchronization process of multi-paralleled grid forming (GFM) inverters is studied detailly. Firstly, based on the multi-paralleled inverters control structure and complementary cluster center of inertia relative motion (CCCOI-RM), the single machine infinite bus (SMIB) system is proposed. Subsequently, the power angle characteristic (PAC) equation of multi-paralleled inverters for different dispatching strategies with simplified damping coefficient is built. Meanwhile, a grid dispatching strategy considering communication delay is proposed, which avoiding the cascading failures of grid faults. Furthermore, according to the grid dispatching strategy, a cluster control strategy optimization method including critical cluster identification (CCI) method and transient stability prediction analysis method is designed, which is proposed to identify unstable cluster in multi-paralleled inverters. In addition, the synchronization stability criteria for dispatching strategies of multi-paralleled inverters are obtained based on the PAC equation, in which the transient instability phenomenon and mechanism for different fault stages is explained by extended equal area criterion (EEAC) method. Finally, the simulations results validate the effectiveness of the theoretical analysis and proposed method.

A simulation study of magnetic nanoparticle clustering in a fluid flow

The assembly of magnetic nanoparticles has been promisingly used in biomedical applications. Several computational studies have addressed the study of cluster formation in the colloidal regime. However, these studies did not mention the formation of clusters in nonequilibrium conditions, such as fluid flow. To fill this gap, we develop computational models of magnetohydrodynamic systems to study their behavior within a flow in terms of cluster shape formation. All simulations were performed by Verlet integration through LAMMPS, a molecular dynamics framework. Through simulations, we discover that the fluid flow promotes formation of non-linear clusters, and linear clusters tend to orient toward the direction of the flow. These phenomena are investigated by using the equilibrium of magnetic torque and fluid torque. The analysis shows that the formation of non-linear clusters are mainly assisted by a bias-orientation of the unstable linear clusters, while the bias-orientation is originated from the torque cancellation in the direction of the flow.

Geo-spatial high-risk clusters of Tuberculosis in the global general population: a systematic review

IntroductionThe objective of this systematic review is to identify tuberculosis (TB) high-risk among the general population globally. The review was conducted using the following steps: elaboration of the research question, search for relevant publications, selection of studies found, data extraction, analysis, and evidence synthesis.MethodsThe studies included were those published in English, from original research, presented findings relevant to tuberculosis high-risk across the globe, published between 2017 and 2023, and were based on geospatial analysis of TB. Two reviewers independently selected the articles and were blinded to each other`s comments. The resultant disagreement was resolved by a third blinded reviewer. For bibliographic search, controlled and free vocabularies that address the question to be investigated were used. The searches were carried out on PubMed, LILACS, EMBASE, Scopus, and Web of Science. and Google Scholar.ResultsA total of 79 published articles with a 40-year study period between 1982 and 2022 were evaluated. Based on the 79 studies, more than 40% of all countries that have carried out geospatial analysis of TB were from Asia, followed by South America with 23%, Africa had about 15%, and others with 2% and 1%. Various maps were used in the various studies and the most used is the thematic map (32%), rate map (26%), map of temporal tendency (20%), and others like the kernel density map (6%). The characteristics of the high-risk and the factors that affect the hotspot’s location are evident through studies related to poor socioeconomic conditions constituting (39%), followed by high population density (17%), climate-related clustering (15%), high-risk spread to neighbouring cities (13%), unstable and non-random cluster (11%).ConclusionThere exist specific high-risk for TB which are areas that are related to low socioeconomic conditions and spectacular weather conditions, these areas when well-known will be easy targets for intervention by policymakers. We recommend that more studies making use of spatial, temporal, and spatiotemporal analysis be carried out to point out territories and populations that are vulnerable to TB.

Clustering evolution during low temperature aging and thermal stability during two-step aging in Al-Mg-Si alloys

The age-hardening response during two-step aging in Al-Mg-Si alloys depends on the nanoclusters formed at the early stage of phase decomposition. However, understanding of the evolution and thermal stability of the nanoclusters is still lacking. This paper discusses clustering behavior during natural aging (NA) and pre-aging (PA) at 50 ℃ through various experiments [hardness, differential scanning calorimetry (DSC), electrical resistivity and three-dimensional atom probe (3DAP)]. A rapid increase in hardness is confirmed due to the acceleration of cluster formation during PA, and a larger area of an endothermic peak is confirmed during DSC thermal analysis compared to NA specimens. Hardness and electrical resistivity were clearly decreased in the PA specimens at the initial stage of two-step aging since the formation of thermally unstable clusters, the Mg-enriched clusters identified by 3DAP, was accelerated during PA. A long aging time at 50 ℃ produces Mg-enriched clusters rather than facilitating the transition from Cluster (1) to Cluster (2). To investigate the atomic arrangement of a cluster, we conducted further analysis for chemical composition of clusters through the normalization for the different sizes of cluster. We found that the Mg-rich clusters with the core-shell structure where Mg atoms are segregated around the cluster-matrix interface, cause thermally unstable at the initial stage of two-step aging. The direct experimental evidence primarily proves that atomic arrangement inside a cluster is a more critical factor than its size for the thermal stability of a cluster in Al-Mg-Si alloys.

Shannon information entropy, soliton clusters and Bose-Einstein condensation in log gravity

We give a probabilistic interpretation of the configurational partition function of the logarithmic sector of critical cosmological topologically massive gravity, in which the Hurwitz numbers considered in our previous works assume the role of probabilities in a distribution on cycles of permutations. In particular, it is shown that the permutations are distributed according to the Ewens sampling formula which plays a major role in the theory of partition structures and their applications to diffusive processes of fragmentation, and in random trees. This new probabilistic result together with the previously established evidence of solitons in the theory provide new insights on the instability originally observed in the theory. We argue that the unstable propagation of a seed soliton at single particle level induces the generation of fragments of defect soliton clusters with rooted tree configuration at multiparticle level, providing a disordered landscape. The Shannon information entropy of the probability distribution is then introduced as a measure of the evolution of the unstable soliton clusters generated. Finally, based on Feynman’s path integral formalism on permutation symmetry in the λ-transition of liquid helium, we argue that the existence of permutation cycles in the configurational log partition function indicates the presence of Bose-Einstein condensates in log gravity.

The multifaceted role of GCM1 during trophoblast differentiation in the human placenta

Remodeling of the uterine vasculature by invasive extravillous trophoblasts (EVTs) is a critical aspect of human placentation. Insufficient EVT invasion can lead to severe obstetrical complications like preeclampsia, intrauterine growth restriction, and preterm birth. Glial cells missing-1 (GCM1) is a transcription factor that is crucial for proper placentation in mice, and is highly expressed in human syncytiotrophoblast (ST) and EVTs. GCM1 is classically considered a master regulator of ST formation, but little is known about its contribution to the development and function of EVTs. Therefore, in this study we test the hypothesis that GCM1 is a critical regulator of both EVT and ST development and function. We show that GCM1 is highly expressed in human trophoblast stem (TS) cells differentiated into either ST or EVTs. Knockdown of GCM1 in TS cells hindered differentiation into both ST and EVT pathways. When placed in ST media, GCM1-knockdown cells formed small, unstable clusters; when placed in EVT media, cells had altered morphology and transcript profiles resembling cells trapped in an intermediate state between CT and EVT, and invasive capacity through matrix was reduced. RNA sequencing analysis of GCM1-deficient TS cells revealed downregulation of EVT-associated genes and enrichment in transcripts related to WNT signaling, which was linked to decreased expression of the EVT master regulator ASCL2 and the WNT antagonist NOTUM. Our findings reveal an essential role of GCM1 during ST and EVT development, and suggest that GCM1 regulates differentiation of human TS cells into EVTs by inducing expression of ASCL2 and NOTUM.

Migration of nitrite corrosion inhibitor in calcium silicate hydrate nanopore: A molecular dynamics simulation study

Nitrite is an effective corrosion inhibitor that can inhibit the corrosion of steel reinforcement and extend the service life of reinforced concrete. The transport speed of nitrite in the cement micro-porous channels determines the anti-corrosion effect of reinforcing steel. In this paper, the transport behavior of three nitrite corrosion inhibitors, namely LiNO2, Ca(NO2)2, and NaNO2, in cement micro-porous channels is investigated based on molecular dynamics simulations and finds that NO2−in LiNO2has the fastest transport speed in the channels. The ions’ local structure and dynamic behavior of ions analysis reveal that ion clusters and pores adsorption determine the speed of ion transport. Compared with the other two nitrites, the Li ions have the weakest ability to capture NO2−and form the most unstable clusters (NO2−-Li-water) in the LiNO2solution. Meanwhile, the coordination numbers results indicate that water in LiNO2provides the most potent driving force. These phenomena reasonably explain the fastest transport speed of NO2−through the pores in LiNO2solution. The transport behavior of corrosion inhibitors in the micro-porous channels is thoroughly decoded at the atomic level, which is instrumental in solving the problem of the optimal corrosion inhibitor selecting for the design of highly durable concrete.

Illuminating History: A Retrospective of Seven Decades

Illuminating History: A Retrospective of Seven Decades

Role of actin filaments and cis binding in cadherin clustering and patterning

Cadherins build up clusters to maintain intercellular contact through trans and cis (lateral) bindings. Meanwhile, interactions between cadherin and the actin cytoskeleton through cadherin/F-actin linkers can affect cadherin dynamics by corralling and tethering cadherin molecules locally. Despite many experimental studies, a quantitative, mechanistic understanding of how cadherin and actin cytoskeleton interactions regulate cadherin clustering does not exist. To address this gap in knowledge, we developed a coarse-grained computational model of cadherin dynamics and their interaction with the actin cortex underlying the cell membrane. Our simulation predictions suggest that weak cis binding affinity between cadherin molecules can facilitate large cluster formation. We also found that cadherin movement inhibition by actin corralling is dependent on the concentration and length of actin filaments. This results in changes in cadherin clustering behaviors, as reflected by differences in cluster size and distribution as well as cadherin monomer trajectory. Strong cadherin/actin binding can enhance trans and cis interactions as well as cadherin clustering. By contrast, with weak cadherin/actin binding affinity, a competition between cadherin-actin binding and cis binding for a limited cadherin pool leads to temporary and unstable cadherin clusters.

K‐means clustering based energy and trust management routing algorithm for mobile ad‐hoc networks

SummaryMANETs (Mobile Ad‐Hoc Networks) have limited energy and are unable to energize themselves. This work aims to create a power management strategy for the MANET that saves energy. While routing based on the cluster, the cluster heads may fail or work improperly by the cause of power issues. As a result, the cluster heads experience instability when gathering information and interacting. The main focus of this research is identifying the unstable cluster heads and replacing them with other nodes that implement the self‐configurable cluster method. For defining the cluster heads successfully, we presented a k‐means protocol approach. The proposed k‐means technique identifies the abnormal or unnecessary variations of the cluster heads and the change in cluster count. A node is designated as a cluster head (CH) in MANET after the cluster value remains lesser than the existing routing threshold value. Information gathering, computing trust levels, and trust‐based configuration are the three key elements of the proposed trust management system, which communicates with each node to establish secure network connections. Finally, the testing results show that the proposed approach uses less energy and effectively manages trust in MANET.

Poisson degree corrected dynamic stochastic block model

Stochastic Block Model (SBM) provides a statistical tool for modeling and clustering network data. In this paper, we propose an extension of this model for discrete-time dynamic networks that takes into account the variability in node degrees, allowing us to model a broader class of networks. We develop a probabilistic model that generates temporal graphs with a dynamic cluster structure and time-dependent degree corrections for each node. Thanks to these degree corrections, the nodes can have variable in- and out-degrees, allowing us to model complex cluster structures as well as interactions that decrease or increase over time. We compare the proposed model to a model without degree correction and highlight its advantages in the case of inhomogenous degree distributions in the clusters and in the recovery of unstable cluster dynamics. We propose an inference procedure based on Variational Expectation-Maximization (VEM) that also provides the means to estimate the time-dependent degree corrections. Extensive experiments on simulated and real datasets confirm the benefits of our approach and show the effectiveness of the proposed algorithm.

Comparative effects of 6-weeks progressive resistance exercise and moderate intensity aerobic exercise on CD4 count and weights of people living with HIV/AIDS in Alex-Ekwueme Federal University Teaching Hospital Ebonyi State.

Introduction:The most significant clinical challenges in people living with HIV/AIDS (PLWHA) are decline in CD4+ T helper cells and abnormal weight reduction. Consequently, PLWHA who are on Anti-Retroviral Therapy (ART) or Highly Active Anti-Retroviral Therapy (HAART) are usually predisposed to coronary artery diseases due to abnormal weight gain (lipohypertrophy), though with improved and unstable Cluster of Differentiation 4 (CD4) counts.The main aim of this study was to compare the effects of 6 weeks progressive resistance exercise (PRE) and moderate intensity aerobic exercise (MIAE) on CD4 count, and weight of PLWHA in Alex-Ekwueme Federal University Teaching Hospital Ebonyi State.Method:The study adopted quasi experimental research design. The population of the study was 60 Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome patients that attended HIV clinics at Alex-Ekwueme Federal University Teaching Hospital Abakaliki (AE-FUTHA) which formed 58 sample sizes for this study.Simple random sampling technique was adopted for the study and flow cytometry, Heel Densitometer (X-rite 331C), and Omron BF 400 were the instrument used for data collection of CD4 counts and weight respectively; and they are standardized hence not validated. Mean, standard deviation and Analysis of Co-variance (ANCOVA) were used to analyze the data obtained. The reliability coefficient obtained from the pilot study was 0.848 and 0.994 for CD4 and WEIGHT respectively.Results:The major findings revealed a significant difference after 6 weeks’ PRE and MIAE on CD4 and Weight of PLWHA.Conclusion:Health promotion policy maker's arm of the government and Healthcare givers should integrate and enforce the use of exercises either as a single regimen or combined exercises into the management plan of PLWHA for greater boost in absolute CD4 count bearing in mind that both have positive effects. Physiotherapist should be integrated in the managements of PLWHA for appropriate prescription and education of therapeutic exercises for body weight.

Endohedral Metallofullerenes: An Ideal Platform of Sub‐Nano Chemistry

Comprehensive SummaryThe hollow cavity of fullerenes is an ideal host for the stabilization of a variety of otherwise unstable metallic clusters, enabling the discovery of novel compounds and unconventional metal‐metal and metal‐carbon interactions. Meanwhile, the internal metallic species exert interesting influences on the chemical properties of the cage carbons, mainly through electronic interactions, leading to unexpectedly high regioselectivity and high yield, and even quantitative conversion. What is the most favorite and original chemistry developed in your research group?Creation of novel metal‐carbon hybrid molecules with unconventional techniques, known as endohedral metallofullerenes, and the unexplored chemistry hidden inside them, especially discovery of the unprecedented bonding between rare earth metal ions and a variety of novel metal clusters, e.g., Ti3C3 and Dy2N4.How do you get into this specific field? Could you please share some experiences with our readers?I entered the field of fullerenes when I was pursuing my Ph.D. during 2001—2004 in Peking University. At that time, fullerenes are no longer a hot research subject as carbon nanotubes which have shown huge perspectives in both fundamentals and applications. However, I like fullerenes because of their clear structures and readily high solubility in organic solvents, which are rather attractive to chemists. I have been working in this field for nearly twenty years and my feeling is that interest and passion are the most important characteristics for a scientist to keep finding in a seemingly out‐of‐date field.What are your hobbies? What are your favorite books?Badminton is my favorite sport in addition to jogging, and reading historical novels and essays is a hobby during my leisure time. Myfavorite books are “Romance of the Three Kindoms”, “War and Peace”, “A Dream of Red Mansions” and “One Hundred Years of Solitude”.How do you supervise your students?As every student has his/her own talents so that I treat each as individuals to express their own merits as much as possible. In addition, the graduate students in my lab do not only learn about the detailed experimental skills but also the way how to find solutions to the scientific problems they are facing, namely thinking modes to solve problems. In detail, I request my students to ensure their time and energy in experiments and research works daily, as diligence is the magic weapon of success, in addition to persistence and enthusiasm.How do you keep balance between research and family?My motivation of hard‐working comes from my family so that I concentrate on my work during working time every day, but I try my best to stay with my kids on weekends and leisure time to replenish my energy.Could you please give us some advices on improving Chinese Journal of Chemistry?This journal should cover all categories of chemistry, keeping balance among organic, inorganic, materials, analytical and others, instead of focusing too much on organic and synthetic chemistry.

Hidden Markov model segmentation to demarcate trajectories of residual apnoea-hypopnoea index in CPAP-treated sleep apnoea patients to personalize follow-up and prevent treatment failure.

Continuous positive airway pressure (CPAP), the reference treatment for obstructive sleep apnoea (OSA), is used by millions of individuals worldwide with remote telemonitoring providing daily information on CPAP usage and efficacy, a currently underused resource. Here, we aimed to implement data science methods to provide tools for personalizing follow-up and preventing treatment failure. We analysed telemonitoring data from adults prescribed CPAP treatment. Our primary objective was to use Hidden Markov models (HMMs) to identify the underlying state of treatment efficacy and enable early detection of deterioration. Secondary goals were to identify clusters of rAHI trajectories which need distinct therapeutic strategies. From telemonitoring records of 2860 CPAP-treated patients (age: 66.31 ± 12.92 years, 69.9% male), HMM estimated three states differing in variability within a given state and probability of shifting from one state to another. The daily inferred state informs on the need for a personalized action, while the sequence of states is a predictive indicator of treatment failure. Six clusters of rAHI trajectories were identified ranging from well-controlled patients (cluster 0: 669 (23%); mean rAHI 0.58 ± 0.59 events/h) to the most unstable (cluster 5: 470 (16%); mean rAHI 9.62 ± 5.62 events/h). CPAP adherence was 30 min higher in cluster 0 compared to clusters 4 and 5 (P value < 0.01). This new approach based on HMM might constitute the backbone for deployment of patient-centred CPAP management improving the personalized interpretation of telemonitoring data, identifying individuals for targeted therapy and preventing treatment failure or abandonment. The online version contains supplementary material available at 10.1007/s13167-021-00264-z.