Data Alignment Research Articles

Recommendation for Clarifying FDA Policy in Evaluating "Sameness" of Higher Order Structure for Generic Peptide Therapeutics.

Recognizing the approach of a dramatic expansion of peptide therapeutics reaching the marketplace in recent years, led by GLP-1 receptor agonists such as semaglutide and liraglutide, the Center for Drug Evaluation and Research (CDER) branch of the US Food and Drug Administration (FDA) issued a final guidance in 2021 that was intended to assist generic drug producers in meeting Abbreviated New Drug Application (ANDA) obligations to establish "sameness" of their active peptide drug relative to that produced by innovator companies. Research and a published report by FDA scientists on best practices followed, which promulgated the use of nuclear magnetic resonance (NMR) and principal component analysis (PCA) and established a quantitative standard by which "sameness" of higher order structure for the applicant's peptide drug could be judged. A key requirement is that drug product samples be analyzed directly and non-invasively, a condition which in practice restricts sample modification to the addition of a small amount of deuterium oxide to allow signal lock and spectral data alignment (as required for NMR analysis). In the study described herein, data are presented to illustrate that 1) relatively small differences in sample pH can cause significant shifting of certain proton resonances, 2) that such resonance shifting is readily reversible and due to the degree of protonation of specific amino acid residues (rather than reflecting differences in higher order structure), and 3) that small differences in pH variability between sample cohorts can frequently cause failure to meet the quantitative benchmark established by the agency. Methodology is presented by which drug sample pHs can be aligned with minimal impact, and a recommendation is made that minor sample pH adjustments be allowed in assessing "sameness" of peptide drug higher order structure.

DiMA: sequence diversity dynamics analyser for viruses.

Sequence diversity is one of the major challenges in the design of diagnostic, prophylactic, and therapeutic interventions against viruses. DiMA is a novel tool that is big data-ready and designed to facilitate the dissection of sequence diversity dynamics for viruses. DiMA stands out from other diversity analysis tools by offering various unique features. DiMA provides a quantitative overview of sequence (DNA/RNA/protein) diversity by use of Shannon's entropy corrected for size bias, applied via a user-defined k-mer sliding window to an input alignment file, and each k-mer position is dissected to various diversity motifs. The motifs are defined based on the probability of distinct sequences at a given k-mer alignment position, whereby an index is the predominant sequence, while all the others are (total) variants to the index. The total variants are sub-classified into the major (most common) variant, minor variants (occurring more than once and of incidence lower than the major), and the unique (singleton) variants. DiMA allows user-defined, sequence metadata enrichment for analyses of the motifs. The application of DiMA was demonstrated for the alignment data of the relatively conserved Spike protein (2,106,985 sequences) of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and the relatively highly diverse pol gene (2637) of the human immunodeficiency virus-1 (HIV-1). The tool is publicly available as a web server (https://dima.bezmialem.edu.tr), as a Python library (via PyPi) and as a command line client (via GitHub).

Definitions of clinical study outcome measures for cardiovascular diseases: the European Unified Registries for Heart Care Evaluation and Randomized Trials (EuroHeart).

Standardized definitions for outcome measures in randomized clinical trials and observational studies are essential for robust and valid evaluation of medical products, interventions, care, and outcomes. The European Unified Registries for Heart Care Evaluation and Randomised Trials (EuroHeart) project of the European Society of Cardiology aimed to create international data standards for cardiovascular clinical study outcome measures. The EuroHeart methods for data standard development were used. From a Global Cardiovascular Outcomes Consortium of 82 experts, five Working Groups were formed to identify and define key outcome measures for: cardiovascular disease (generic outcomes), acute coronary syndrome and percutaneous coronary intervention (ACS/PCI), atrial fibrillation (AF), heart failure (HF) and transcatheter aortic valve implantation (TAVI). A systematic review of the literature informed a modified Delphi method to reach consensus on a final set of variables. For each variable, the Working Group provided a definition and categorized the variable as mandatory (Level 1) or optional (Level 2) based on its clinical importance and feasibility. Across the five domains, 24 Level 1 (generic: 5, ACS/PCI: 8, AF: 2; HF: 5, TAVI: 4) and 48 Level 2 (generic: 18, ACS-PCI: 7, AF: 6, HF: 2, TAVI: 15) outcome measures were defined. Internationally derived and endorsed definitions for outcome measures for a range of common cardiovascular diseases and interventions are presented. These may be used for data alignment to enable high-quality observational and randomized clinical research, audit, and quality improvement for patient benefit.

Automatic Conditioning of Marine Seismic Angle Stack Data With a Convolutional Neural Network

Misalignment of reflections is a common problem in migrated prestack seismic data, which occurs due to moveout correction with incomplete knowledge of the velocity model. This issue persists in partial angle stack data often used for amplitude versus offset (AVO) analyses, and is traditionally treated with residual moveout and trim static corrections. However, these methods require time-consuming velocity selection and parameter tuning, and commonly lead to spurious alignment of reflections with noise. Therefore, we train a convolutional neural network (CNN) on synthetic examples to automatically perform conditioning and alignment of angle stack data. First, two-dimensional synthetic common depth point (CDP) gathers are created using a convolutional modeling method, and then made into input-target pairs where the inputs are poorly moveout-corrected (using inaccurate normal moveout velocities) and the targets are accurately moveout-corrected. These input-target examples are split into angle stacks for the near, mid, and far offsets, and then used to train a CNN to transform the misaligned angle stacks into their more-aligned counterparts. In testing, the trained network increases the alignment of unseen synthetic data, improving the correlation with the target far offset trace from 0.76 to 0.85 on average. The network is applied on two field examples from offshore Norway with Class 3 and Class 2P AVO responses, respectively. In both cases, the angle stack data predicted by the network shows improved alignment, greater resolution in the far offset stacked section, and increased correspondence to a well-tie synthetic seismogram, all while retaining the AVO responses. The CNN predictions are compared to angle stack data following traditional conditioning (residual moveout and trim static corrections), with the results being of similar or greater quality. This method enables high quality conditioning of angle stack data at a fraction of the time required for conventional methods, and is easily adaptable to different datasets.

Management information systems and organizational agility: a bibliometric analysis

Purpose This paper aims to complex relationship between management information systems (MIS) and organizational agility, identifying gaps and key themes to guide future research in this domain. Design/methodology/approach A systematic extraction of 578 relevant articles from the Scopus database was conducted to provide an in-depth bibliometric analysis of the evolving role of MIS in enhancing organizational agility. Data analysis and visualization were performed using R Studio and VOSviewer. Findings The analysis highlights significant publication trends and identifies leading countries, institutions and journals in MIS and organizational agility research. Collaborative efforts from nations such as the USA, China and the UK, were prominent, focusing on information technology alignment, strategic agility and big data analytics. This study delineates distinct research clusters and future research questions, offering a clear trajectory for ongoing exploration in the field. Originality/value To the best of the authors’ knowledge, this study is one of the first to provide a comprehensive bibliometric analysis of the intersection between MIS and organizational agility. It offers valuable insights into research trends and future directions while emphasizing the need for MIS deployment, thereby contributing to a comprehensive understanding of the field.

Towards practical data alignment in production federated learning

Towards practical data alignment in production federated learning

A numerical evaluation of real-time workloads for ramp controller through optimization of multi-type feature combinations derived from eye tracker, respiratory, and fatigue patterns.

Ramp controllers are required to manage their workloads effectively while handling complex operational tasks, a crucial part of improving aviation safety. The ability to detect their instantaneous workload is vital for ensuring operational effectiveness and preventing hazardous incidents. This paper introduces a novel methodology aimed at enhancing the evaluation of the ramp controller's cumulative workload by incorporating and optimizing the feature combination from eye movement, respiratory, and fatigue characteristics. Specifically, a 90-minute simulated experiment related to ramp control tasks, using real data from Shanghai Hongqiao Airport, is conducted to collect multi-type data from 8 controllers. Following data construction and the extraction of multi-type, the workloads of all samples are categorized through unsupervised learning. Subsequently, supervised learning techniques are used to calculate feature weights and train classifiers after data alignment. The optimal feature combination is established by calculating feature weights, and the best classification accuracy is over 98%, achieved by the KNN classifier. Furthermore, numerical evaluation and threshold calculations for different workload levels are interpreted. It is promising to provide insights into future works towards human-centered data construction, processing, and interpretation to promote the progress of workload assessment within the aviation industry.

Application of computational algorithms for single-cell RNA-seq and ATAC-seq in neurodegenerative diseases.

Recent advancements in single-cell technologies, including single-cell RNA sequencing (scRNA-seq) and Assay for Transposase-Accessible Chromatin using sequencing (scATAC-seq), have greatly improved our insight into the epigenomic landscapes across various biological contexts and diseases. This paper reviews key computational tools and machine learning approaches that integrate scRNA-seq and scATAC-seq data to facilitate the alignment of transcriptomic data with chromatin accessibility profiles. Applying these integrated single-cell technologies in neurodegenerative diseases, such as Alzheimer's disease and Parkinson's disease, reveals how changes in chromatin accessibility and gene expression can illuminate pathogenic mechanisms and identify potential therapeutic targets. Despite facing challenges like data sparsity and computational demands, ongoing enhancements in scATAC-seq and scRNA-seq technologies, along with better analytical methods, continue to expand their applications. These advancements promise to revolutionize our approach to medical research and clinical diagnostics, offering a comprehensive view of cellular function and disease pathology.

Data Organization Patterns in Data Mesh: Optimizing Data Management for the Modern Enterprise

This article explores the implementation of Data Mesh principles through a structured layered approach to data organization, addressing the challenges of managing large-scale, diverse data ecosystems in modern enterprises. As global data creation surges towards 181 zettabytes by 2025, traditional centralized data architectures are proving inadequate. We propose a four-layer methodology—Raw, Work, Gold, and Publish—that aligns with Data Mesh philosophy while providing clear delineation of data processing stages. Through case studies across various industries, we demonstrate how this approach leads to significant improvements in data quality, time-to-insight, and incident resolution. A large financial institution achieved a 40% reduction in time-to-insight, while a global manufacturer reported a 30% improvement in data quality. The article also discusses the challenges of implementation, including cultural shifts and skill gaps. Our findings suggest that combining Data Mesh principles with a layered organizational structure can transform enterprise data landscapes, enhancing agility, cross-functional collaboration, and alignment of data assets with business objectives. This article contributes to the evolving field of decentralized data architectures and provides practical insights for organizations seeking to optimize their data management strategies in the face of exponential data growth.

Focal mechanism determination by location-constrained deep learning: application to microseismic monitoring

Accurate and rapid determination of focal mechanism solutions is of great significance for real-time seismic monitoring. Previous deep learning approaches for focal mechanism determination typically use the waveform data, which is sensitive to the velocity model and inherently includes information about both the source location and the focal mechanism. We introduce a location-constrained deep learning algorithm for determining the focal mechanism for surface microseismic events. By using aligned P-wave data along with azimuth and take-off angle as input, we narrow the solution space for the focal mechanism problem and reduce the dependence on the velocity model. The model is trained using 40,000 theoretical samples generated with the geometry and velocity model of the field data. Validation tests, comparisons with a waveform-based network, velocity perturbation tests, and location error tests are performed to demonstrate the robustness and efficiency of our method. After applying the trained model to field data, the results demonstrate that our method is fast and achieves comparable accuracy to HASH results for high-quality events, making it promising for real-time microseismic monitoring.

Feature Transfer Learning for Fatigue Life Prediction of Additive Manufactured Metals With Small Samples

ABSTRACTA feature transfer learning (FTL)‐based model is proposed to address small‐sample problems in fatigue life prediction of additively manufactured (AM) metals. Transfer component analysis (TCA) is studied for data alignment before model training. Correspondingly, two TCA improvement strategies are further considered to aggregate training data from distinct AM processing conditions. An experimental database consisting of 103 fatigue data is built for model evaluation. The results demonstrate that the proposed model outperforms conventional machine learning models and other transfer learning‐based models in terms of accuracy and data demand, showing good applicability for AM fatigue life assessment.

Pharmacologic prophylaxis of postoperative delirium in elderly patients: A network meta-analysis of randomized controlled trials

BackgroundThe high incidence and mortality rates of postoperative delirium (POD) among elderly patients highlights the pressing need for tailored prophylactic strategies. Despite various pharmacologic prophylactic strategies have been reported effective, their overall benefit and safety remain unclear in the geriatric population. Our network meta-analysis (NMA) aimed to systematically evaluate and rank the effectiveness of various pharmacological interventions in preventing POD in elderly patients. MethodsWe conducted an extensive search of PubMed, Embase, Cochrane Central Register of Controlled Trials, PsycINFO, and Google Scholar for randomized controlled trials (RCTs) published up to August 1, 2023. We included RCTs examining pharmacological prophylactic effects of POD in elderly patients. To extract data in alignment with predefined areas of interest, we employed the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. The primary outcome was the incidence of POD. For secondary outcomes, we evaluated tolerability through all-cause discontinuation or drop-out rates, as well as all-cause mortality. ResultsOur analysis encompassed a total of 44 RCTs involving 11,178 patients. Out of these, 26 RCTs involved comparisons with placebo only. For delirium prevention, the treatment groups receiving atypical antipsychotics (odds ratio (OR) of 0.27 and 95% confidence interval (CI) of 0.12-0.58), haloperidol (OR of 0.42; 95% CI of 0.25-0.71), dexmedetomidine (OR of 0.51 and 95% CI of 0.37-0.71 and melatonergic agents (MMA) (OR of 0.57 and 95% CI of 0.33-0.98) had significantly lower rates of delirium compared to the placebo group. Notably, the atypical antipsychotics ranked as the most effective treatment. For tolerability, no statistically differences in rates of dropout discontinuation and all-cause mortality among groups allocated to the placebo or individual pharmacological treatments. ConclusionsBased on indirect evidence, our network meta-analysis identified atypical antipsychotics, dexmedetomidine, MMA, and haloperidol as effective in preventing POD in the elderly, with atypical antipsychotics ranking highest. However, it is essential to note that these findings should be confirmed through further RCTs.

Translation of MFL and UT data by using generative adversarial networks: A comparative study

Magnetic flux leakage (MFL) and ultrasonic testing (UT) are widely used in-line inspection technologies to detect corrosion defects along pipelines. The integration of MFL and UT data has the potential to provide complementary insights that facilitate a comprehensive assessment of pipeline integrity. However, due to the inherent dissimilarity with their underlying physical principles, these techniques yield notable disparities in signal characteristics, posing challenges in integrating these multimodal data. This study aims to establish a translation mapping between MFL and UT signals to achieve consistent physical interpretations across the two modalities. Thus, this study explored the feasibility of generative adversarial network (GAN) based models encompassing both supervised and unsupervised translation approaches contingent on the availability of aligned data. Furthermore, two translation modes, MFL-UT and UT-MFL, were analyzed separately to understand the effectiveness of the translation direction. The experimental results demonstrate satisfactory performance for both aligned and unaligned data translation, with the UT-MFL translation direction yielding superior results. Overall, the translation approaches pave the way for future applications, especially in subsequent data analysis tasks such as registration, comparison, and fusion of multimodal data.

Mitochondrial genome insights into the spatio-temporal distribution and genetic diversity of Dendrobium hancockii Rolfe (Orchidaceae).

With its distinctive evolutionary rate and inheritance patterns separate from the nuclear genome, mitochondrial genome analysis has become a prominent focus of current research. Dendrobium hancockii Rolfe, a species of orchid with both medicinal and horticultural value, will benefit from the application of the fully assembled and annotated mitochondrial genome. This will aid in elucidating its phylogenetic relationships, comparative genomics, and population genetic diversity. Based on sequencing results from Illumina combined with PacBio and Nanopore, the mitochondrial genome map of D. hancockii was constructed. Comparative analysis was conducted from the perspectives of phylogeny across multiple species, selection pressure on protein-coding genes, and homologous segments. The population diversity of D. hancockii was analyzed using single nucleotide polymorphism (SNP) data from the mitochondrial genome and single-copy nuclear genes. This research constructed a circular mitochondrial map for D. hancockii, spanning 523,952 bp, containing 40 unique protein-coding genes, 37 transfer RNA genes, and 4 ribosomal RNA genes. Comparative analysis of mitochondrial genes from 26 land plants revealed a conserved gene cluster, "rpl16-ccmFn-rps3-rps19," particularly within the Dendrobium genus. The mitochondrial genome of D. hancockii exhibits a lower point mutation rate but significant structural variation. Analysis of 103 resequencing samples identified 19,101 SNP sites, dividing D. hancockii into two major groups with limited gene flow between them, as supported by population diversity, genetic structure analysis, principal component analysis, and phylogenetic trees. The geographical distribution and genetic differentiation of D. hancockii into two major groups suggest a clear phytogeographical division, likely driven by ancient geological or climatic events. The close alignment of mitochondrial data with nuclear gene data highlights the potential of the mitochondrial genome for future studies on genetic evolution in this species.

Automated volumetric analysis of the inner ear fluid space from hydrops magnetic resonance imaging using 3D neural networks

Due to the development of magnetic resonance (MR) imaging processing technology, image-based identification of endolymphatic hydrops (EH) has played an important role in understanding inner ear illnesses, such as Meniere’s disease or fluctuating sensorineural hearing loss. We segmented the inner ear, consisting of the cochlea, vestibule, and semicircular canals, using a 3D-based deep neural network model for accurate and automated EH volume ratio calculations. We built a dataset of MR cisternography (MRC) and HYDROPS-Mi2 stacks labeled with the segmentation of the perilymph fluid space and endolymph fluid space of the inner ear to devise a 3D segmentation deep neural network model. End-to-end learning was used to segment the perilymph fluid and the endolymph fluid spaces simultaneously using aligned pair data of the MRC and HYDROPS-Mi2 stacks. Consequently, the segmentation performance of the total fluid space and endolymph fluid space had Dice similarity coefficients of 0.9574 and 0.9186, respectively. In addition, the EH volume ratio calculated by experienced otologists and the EH volume ratio value predicted by the proposed deep learning model showed high agreement according to the interclass correlation coefficient (ICC) and Bland–Altman plot analysis.

The Framework of GeoSOT-3D Grid Modeling for Spatial Artificial Intelligence

Abstract. With the development of society, spatial artificial intelligence (spatial AI) research is gradually able to play a greater role. However, spatial AI has problems such as data alignment, poor interpretability, and cross domain learning. Therefore, this paper proposes an innovative GeoSOT-3D grid modeling framework for spatial AI research, which enhances the application capabilities of spatial AI. Grid modeling will be able to run through the upstream and downstream of spatial AI research, providing encoding calculations and spatial neighborhood embedding matrices for spatial data. This paper also uses task examples to demonstrate how to effectively organize and index spatial data using GeoSOT-3D grids and conduct spatial AI research. The use of GeoSOT-3D grids for spatial AI analysis has enormous potential and broad application prospects, which will help promote the further development and application of spatial AI.

Enhancing underwater target detection: Fusion of spatio‐temporal incompletely‐aligned AIS and sonar information via DTW and multi‐head attention mechanism

AbstractIn the field of underwater target detection, the passive sonar is an important means of long‐distance target detection. The sonar detection information typically includes both surface and underwater targets, whereas it is a great challenge on effectively distinguishing between surface and underwater targets solely based on sonar information. Effective fusion of sonar and AIS (Automatic Identification System) data can leverage their complementary nature to compensate for the limitation of sonar information. However, the sonar information and AIS information are acquired based on different detection principles and systems, which are essentially multi‐source heterogeneous information with obvious spatio‐temporal misalignment in nature. Existing fusion methods normally struggle to effectively align sonar and AIS data in both time and space subject to the complexity of the problem. In this study, the Dynamic Time Warping (DTW) algorithm is applied to align sonar and AIS data in the time domain. In addition, a deep learning algorithm with multi‐head attention mechanism is proposed to achieve the spatial alignment of sonar and AIS data, where the matching between the surface targets in AIS data and the same surface targets in sonar data can also be successfully achieved. It provides a priori knowledge to enhance the underwater target detection of the passive sonar by eliminating the interference of the surface targets. Based on the attention mechanism, the abstract features extracted from the intermediate‐layer of the neural networks are found to be effective to represent the typical features of the target motion trajectories, which also demonstrates the effectiveness of the attention mechanism. The experiment results show that the proposed method can successfully achieve a MatchingSucccessRate of over 95% between the AIS targets and sonar detection targets.

Retrospective analysis of COVID-19 clinical and laboratory data: Constructing a multivariable model across different comorbidities

BackgroundThe clinical pathogenesis of COVID-19 necessitates a comprehensive and homogeneous study to understand the disease mechanisms. Identifying clinical symptoms and laboratory parameters as key predictors can guide prognosis and inform effective treatment strategies. This study analyzed comorbidities and laboratory metrics to predict COVID-19 mortality using a homogeneous model. MethodA retrospective cohort study was conducted on 7500 COVID-19 patients admitted to Rasoul Akram Hospital between 2022 and 2022. Clinical and laboratory data, along with comorbidity information, were collected and analyzed using advanced coding, data alignment, and regression analyses. Machine learning algorithms were employed to identify relevant features and calculate predictive probability scores. ResultsThe frequency and mortality rates of COVID-19 among males (19.3 %) were higher than those among females (17 %) (p = 0.01, OR = 0.85, 95 % CI = 0.76–0.96). Cancer (p < 0.05, OR = 1.9, 95 % CI = 1.48–2.4) and Alzheimer's (p < 0.05, OR = 2.36, 95 % CI = 1.89–2.9) were the two most common comorbidities associated with long-term hospitalization (LTH). Kidney disease (KD) was identified as the most lethal comorbidity (45 % of KD patients) (OR = 5.6, 95 % CI = 5.05–6.04, p < 0.001). Age > 55 was the most predictive parameter for mortality (p < 0.001, OR = 6.5, 95 % CI = 1.03–1.04), and the CT scan score showed no predictive value for death (p > 0.05). WBC, Cr, CRP, ALP, and VBG-HCO3 were the most significant critical data associated with death prediction across all comorbidities (p < 0.05). ConclusionCOVID-19 is particularly lethal for elderly adults; thus, age plays a crucial role in disease prognosis. Regarding death prediction, various comorbidities rank differently, with KD having a significant impact on mortality outcomes.

Audio feature enhancement based on quaternion filtering and deep hashing

This paper aims to solve the problems of difficult convergence of audio model training, large data demand, and large dimensionality of storage space for audio-generated feature vectors. To this end, this paper proposes the use of quaternion Gabor filtering to suppress the background information of the spectrogram and reduce the interference of the data for the case of insufficient data alignment between audio data and image data after shifting the domain. In addition, different scales of window lengths and frame shifts are used to capture the connections between different vocal objects. To address the problem that the generated feature vectors are large dimensional, we use a deep hash module to map high-dimensional features to low-dimensional features and use a probability function to make the learned samples more consistent with the overall distribution. In the experimental evaluation, the proposed method was evaluated on the environmental sound classification dataset and the music genre classification dataset. The proposed method uses only a common backbone network and improves the accuracy of audio recognition.

Developing and piloting a set of quality-of-care indicators for Romanian public hospitals as part of a national programme to fund quality

BackgroundHealthcare systems aim to enhance the health status and well-being of the individuals and populations they serve. To achieve this, measuring and evaluating the quality and safety of services provided and the outcomes achieved is essential. Like other countries, Romania faces challenges regarding the quality of care provided in its public hospitals. To address this, the Romanian Ministry of Health initiated reforms in 2022, including implementing a pay-for-performance model based on quality indicators. This paper presents a descriptive analysis of processes, methods, results and lessons learned from developing and piloting a set of Quality of Care indicators for Romanian public hospitals.MethodsWorld Health Organization’s Athens Office on Quality of Care and Patient Safety assisted Romania in developing and piloting a set of quality-of-care indicators for public hospitals. The development phase included defining indicator domains, identifying potential indicators across these domains, and defining the final indicator set. The piloting phase involved selecting and recruiting piloting hospitals, developing data collection and validation methods and tools, training hospital staff, and collecting and analysing indicator data. Piloting ended with an evaluation workshop. Mixed, quantitative and qualitative methods were used, including literature reviews, stakeholder consultation workshops, survey instruments developed for this study, modified Delphi panels and consensus-building meetings. National stakeholders were actively involved throughout the process.ResultsFour priority domains were defined for quality-of-care indicators for Romanian public hospitals: patient safety, patient experience, healthcare workforce training and safety, and clinical effectiveness. 25 core indicators were selected across these domains. During the pilot, hospitals achieved an average completion rate of 90% for data submission, with all domains rated equally relevant during post-pilot evaluations. Lessons included the need for supportive legislation, improved internal auditing practices and enhanced staff training, refinement of indicator data collection methods and alignment of indicators with hospital-specific contexts.ConclusionsThis work presents a significant stride in improving Romanian public hospitals’ quality of care and patient safety. It underscores the importance of high-level commitment, stakeholder engagement, and robust data practices in driving successful quality improvement efforts. Emphasising the role of data-driven and patient-centric approaches in achieving optimal healthcare outcomes, lessons learned offer insights for the continuation of quality improvement work in Romania but also for healthcare systems elsewhere.