Umeå University's logo

Objective: In this paper, we implement and validate a set of optimization approaches that were applied on ambulance data from the Västerbotten county in Sweden collected 2018, with the objective to find the optimal placement of the ambulance stations (or stand-by positions) in Umeå, a municipality in the county with regards to median of response times for priority 1 alarms, the most urgent type of alarms (MRT1).

Methods: Here, we use data-driven approaches for optimizing the placement of ambulance stations. For a given allocation, i.e. placement of the stations, a large-scale simulation is conducted to estimate the allocation's MRT1. Since the inherent mechanism of the simulation function is very complex, the optimization problem has a black-box nature. We use two methods belonging to important classes for solving the problem of black-box optimization: GPS (smooth-free) and surrogate (smooth-based) methods. Both methods can be used on either local or global data and implemented using a one-by-one approach or an all-together approach. To study the mentioned methods and approaches, we consider several real-world scenarios pertaining to the placement of ambulance stations in Umeå municipality.

Results: Relocating the ambulance stations in Umeå can reduce MRT1 around 80-100 seconds in comparison with the current allocation. Using global data leads to better solutions with lower MRT1-values, although they demand more computational time. The results of GPS and surrogate methods are similar, but the surrogate method is less sensitive to the starting position. One-by-one approach is more effective and less time-consuming than the all-together approach.

Conclusion: The results confirm that relocating ambulance stations can lead to a significant decrease in MRT1 and it also can compensate for the loss of an ambulance resource partially. To reduce the dimensionality and the cost of optimization methods, it can be better to use one-by-one approach than all-together.

Background: A multidrug-resistant lineage of Staphylococcus epidermidis named ST215 is a common cause of prosthetic joint infections and other deep surgical site infections in Northern Europe, but is not present elsewhere. The increasing resistance among S. epidermidis strains is a global concern. We used whole-genome sequencing to characterize ST215 from healthcare settings.

Results: We completed the genome of a ST215 isolate from a Swedish hospital using short and long reads, resulting in a circular 2,676,787 bp chromosome and a 2,326 bp plasmid. The new ST215 genome was placed in phylogenetic context using 1,361 finished public S. epidermidis reference genomes. We generated 10 additional short-read ST215 genomes and 11 short-read genomes of ST2, which is another common multidrug-resistant lineage at the same hospital. We studied recombination’s role in the evolution of ST2 and ST215, and found multiple recombination events averaging 30–50 kb. By comparing the results of antimicrobial susceptibility testing for 31 antimicrobial drugs with the genome content encoding antimicrobial resistance in the ST215 and ST2 isolates, we found highly similar resistance traits between the isolates, with 22 resistance genes being shared between all the ST215 and ST2 genomes. The ST215 genome contained 29 genes that were historically identified as virulence genes of S. epidermidis ST2. We established that in the nucleotide sequence stretches identified as recombination events, virulence genes were overrepresented in ST215, while antibiotic resistance genes were overrepresented in ST2.

Conclusions: This study features the extensive antibiotic resistance and virulence gene content in ST215 genomes. ST215 and ST2 lineages have similarly evolved, acquiring resistance and virulence through genomic recombination. The results highlight the threat of new multidrug-resistant S. epidermidis lineages emerging in healthcare settings.

Common objectives in machine learning research are to predict the output quality of manufacturing processes, to perform root cause analysis in case of reduced quality, and to propose intervention strategies. The cost of reduced quality must be weighed against the cost of the interventions, which depend on required downtime, personnel costs, and material costs. Furthermore, there is a risk of false negatives, i.e., failure to identify the true root causes, or false positives, i.e., adjustments that further reduce the quality. A policy for process adjustments describes when and where to perform interventions, and we say that a policy is worthwhile if it reduces the expected operational cost. In this paper, we describe a data-driven alarm and root cause analysis framework, that given a predictive and explanatory model trained on high-dimensional process and quality data, can be used to search for a worthwhile adjustment policy. The framework was evaluated on large-scale simulated process and quality data. We find that worthwhile adjustment policies can be derived also for problems with a large number of explanatory variables. Interestingly, the performance of the adjustment policies is almost exclusively driven by the quality of the model fits. Based on these results, we discuss key areas of future research, and how worthwhile adjustment policies can be implemented in real world applications.

Introduction: The need for innovative technology in healthcare is apparent due to challenges posed by the lack of resources. This study investigates the adoption of AI-based systems, specifically within the postanesthesia care unit (PACU). The aim of the study was to explore staff needs and expectations concerning the development and implementation of a digital patient flow system based on ML predictions.

Methods: A qualitative approach was employed, gathering insights through interviews with 20 healthcare professionals, including nurse managers and staff involved in planning patient flows and patient care. The interview data were analyzed using reflexive thematic analysis, following steps of data familiarization, coding, and theme generation. The resulting themes were then assessed for their alignment with the modified technology acceptance model (TAM2).

Results: The respondents discussed the benefits and drawbacks of the proposed ML system versus current manual planning. They emphasized the need for controlling PACU throughput and expected the ML system to improve the length of stay predictions and provide a comprehensive patient flow overview for staff. Prioritizing the patient was deemed important, with the ML system potentially allowing for more patient interaction time. However, concerns were raised regarding potential breaches of patient confidentiality in the new ML system. The respondents suggested new communication strategies might emerge with effective digital information use, possibly freeing up time for more human interaction. While most respondents were optimistic about adapting to the new technology, they recognized not all colleagues might be as convinced.

Conclusion: This study showed that respondents were largely favorable toward implementing the proposed ML system, highlighting the critical role of nurse managers in patient workflow and safety, and noting that digitization could offer substantial assistance. Furthermore, the findings underscore the importance of strong leadership and effective communication as key factors for the successful implementation of such systems.

Motivated by the development of optimal dispatching strategies for prehospital resources, we model the spatial distribution of ambulance call events in the Swedish municipality Skellefteå during 2014–2018 in order to identify important spatial covariates and discern hotspot regions. Our large-scale multivariate data point pattern of call events consists of spatial locations and marks containing the associated priority levels and sex labels. The covariates used are related to road network coverage, population density, and socio-economic status. For each marginal point pattern, we model the associated intensity function by means of a log-linear function of the covariates and their interaction terms, in combination with lasso-like elastic-net regularized composite/Poisson process likelihood estimation. This enables variable selection and collinearity adjustment as well as reduction of variance inflation from overfitting and bias from underfitting. To incorporate mobility adjustment, reflecting people’s movement patterns, we also include a nonparametric (kernel) intensity estimate as an additional covariate. The kernel intensity estimation performed here exploits a new heuristic bandwidth selection algorithm. We discover that hotspot regions occur along dense parts of the road network. A mean absolute error evaluation of the fitted model indicates that it is suitable for designing prehospital resource dispatching strategies. Supplementary materials accompanying this paper appear online.

Although BCG has been used for almost 100 years to immunize against Mycobacterium tuberculosis, TB remains a global public health threat. Numerous clinical trials are underway studying novel vaccine candidates and strategies to improve or replace BCG, but vaccine development still lacks a well-defined set of immune correlates to predict vaccine-induced protection against tuberculosis. This study aimed to address this gap by examining transcriptional responses to BCG vaccination in C57BL/6 inbred mice, coupled with protection studies using Diversity Outbred mice. We evaluated relative gene expression in blood obtained from vaccinated mice, because blood is easily accessible, and data can be translated to human studies. We first determined that the average peak time after vaccination is 14 days for gene expression of a small subset of immune-related genes in inbred mice. We then performed global transcriptomic analyses using whole blood samples obtained two weeks after mice were vaccinated with BCG. Using comparative bioinformatic analyses and qRT-PCR validation, we developed a working correlate panel of 18 genes that were highly correlated with administration of BCG but not heat-killed BCG. We then tested this gene panel using BCG-vaccinated Diversity Outbred mice and revealed associations between the expression of a subset of genes and disease outcomes after aerosol challenge with M. tuberculosis. These data therefore demonstrate that blood-based transcriptional immune correlates measured within a few weeks after vaccination can be derived to predict protection against M. tuberculosis, even in outbred populations.

Prediction methods can be augmented by local explanation methods (LEMs) to perform root cause analysis for individual observations. But while most recent research on LEMs focus on low-dimensional problems, real-world datasets commonly have hundreds or thousands of variables. Here, we investigate how LEMs perform for high-dimensional industrial applications. Seven prediction methods (penalized logistic regression, LASSO, gradient boosting, random forest and support vector machines) and three LEMs (TreeExplainer, Kernel SHAP, and the conditional normal sampling importance (CNSI)) were combined into twelve explanation approaches. These approaches were used to compute explanations for simulated data, and real-world industrial data with simulated responses. The approaches were ranked by how well they predicted the contributions according to the true models. For the simulation experiment, the generalized linear methods provided best explanations, while gradient boosting with either TreeExplainer or CNSI, or random forest with CNSI were robust for all relationships. For the real-world experiment, TreeExplainer performed similarly, while the explanations from CNSI were significantly worse. The generalized linear models were fastest, followed by TreeExplainer, while CNSI and Kernel SHAP required several orders of magnitude more computation time. In conclusion, local explanations can be computed for high-dimensional data, but the choice of statistical tools is crucial.

Exposing sires to various environmental manipulations has demonstrated that paternal effects can be non-trivial also in species where male investment in offspring is almost exclusively limited to sperm. Whether paternal effects also have a genetic component (i.e. paternal indirect genetic effects (PIGEs)) in such species is however largely unknown, primarily because of methodological difficulties separating indirect from direct effects of genes. PIGEs may nevertheless be important since they have the capacity to contribute to evolutionary change. Here we use Drosophila genetics to construct a breeding design that allows testing nearly complete haploid genomes (more than 99%) for PIGEs. Using this technique, we estimate the variance in male lifespan due to PIGEs among four populations and compare this to the total paternal genetic variance (the sum of paternal indirect and direct genetic effects). Our results indicate that a substantial part of the total paternal genetic variance results from PIGEs. A screen of 38 haploid genomes, randomly sampled from a single population, suggests that PIGEs also influence variation in lifespan within populations. Collectively, our results demonstrate that PIGEs may constitute an underappreciated source of phenotypic variation.

Data on toxic effects are at large missing the prevailing understanding of the risks of industrial chemicals. Thyroid hormone (TH) system disruption includes interferences of the life cycle of the thyroid hormones and may occur in various organs. In the current study, high-throughput screening data available for 14 putative molecular initiating events of adverse outcome pathways, related to disruption of the TH system, were used to develop 19 in silico models for identification of potential thyroid hormone system-disrupting chemicals. The conformal prediction framework with the underlying Random Forest was used as a wrapper for the models allowing for setting the desired confidence level and controlling the error rate of predictions. The trained models were then applied to two different databases: (i) an in-house database comprising xenobiotics identified in human blood and ii) currently used chemicals registered in the Swedish Product Register, which have been predicted to have a high exposure index to consumers. The application of these models showed that among currently used chemicals, fewer were overall predicted as active compared to chemicals identified in human blood. Chemicals of specific concern for TH disruption were identified from both databases based on their predicted activity.

Cancer subtype identification is important to facilitate cancer diagnosis and select effective treatments. Clustering of cancer patients based on high-dimensional RNA-sequencing data can be used to detect novel subtypes, but only a subset of the features (e.g., genes) contains information related to the cancer subtype. Therefore, it is reasonable to assume that the clustering should be based on a set of carefully selected features rather than all features. Several feature selection methods have been proposed, but how and when to use these methods are still poorly understood. Thirteen feature selection methods were evaluated on four human cancer data sets, all with known subtypes (gold standards), which were only used for evaluation. The methods were characterized by considering mean expression and standard deviation (SD) of the selected genes, the overlap with other methods and their clustering performance, obtained comparing the clustering result with the gold standard using the adjusted Rand index (ARI). The results were compared to a supervised approach as a positive control and two negative controls in which either a random selection of genes or all genes were included. For all data sets, the best feature selection approach outperformed the negative control and for two data sets the gain was substantial with ARI increasing from (−0.01, 0.39) to (0.66, 0.72), respectively. No feature selection method completely outperformed the others but using the dip-rest statistic to select 1000 genes was overall a good choice. The commonly used approach, where genes with the highest SDs are selected, did not perform well in our study.