Glycemic control in children with type 1 diabetes treated with the advanced hybrid closed loop system 2-year prospective, observational, two-center study.

Background and aims: MiniMed 780G is the first Advanced Hybrid Closed Loop (AHCL) system in Poland, approved in the EU in 2020. To date, observations of glycemic control up to 12 months have been published. This study aimed to analyze glycemic control and anthropometric parameters in children and adolescents with type 1 diabetes (T1D) after two years of using the AHCL system. Materials and methods: We prospectively collected anthropometric data, pump, and continuous glucose records of fifty T1D children (9.9 ± 2.4 years, 24 (48%) boys, T1D for 3.9 ± 2.56 years) using an AHCL system. We compared the two-week AHCL records obtained after AHCL enrollment with data 6, 12, and 24 months after starting AHCL. Results: Time in range (70-180 mg/dl) and BMI z-score did not change during the 2 years of observation (p>0.05). The percentage of autocorrection in total daily insulin increased significantly (p<0.005). Conclusion: Glycemic control in the investigated group of children with T1D treated with the AHCL system for 2 years remained stable. Children in this group maintained weight and optimal metabolic control, most likely due to autocorrection boluses.

Epigenetic signature of ionizing radiation in therapy-related AML patients.

Therapy-related acute myeloid leukaemia (t-AML) is a late side effect of previous chemotherapy (ct-AML) and/or radiotherapy (rt-AML) or immunosuppressive treatment. t-AMLs, which account for ∼10-20 % of all AML cases, are extremely aggressive and have a poor prognosis compared to de novo AML. Our hypothesis is that exposure to radiation causes genome-wide epigenetic changes in rt-AML. An epigenome-wide association study was undertaken, measuring over 850K methylation sites across the genome from fifteen donors (five healthy, five de novo, and five t-AMLs). The study predominantly focussed on 94K sites that lie in CpG-rich gene promoter regions. Genome-wide hypomethylation was discovered in AML, primarily in intergenic regions. Additionally, genes specific to AML were identified with promoter hypermethylation. A two-step validation was conducted, both internally, using pyrosequencing to measure methylation levels in specific regions across fifteen primary samples, and externally, with an additional eight AML samples. We demonstrated that the MEST and GATA5 gene promoters, which were previously identified as tumour suppressors, were noticeably hypermethylated in rt-AML, as opposed to other subtypes of AML and control samples. These may indicate the epigenetic involvement in the development of rt-AML at the molecular level and could serve as potential targets for drug therapy in rt-AML.

Transcriptional Inflammatory Signature in Healthy Donors and Different Radiotherapy Cancer Patients.

Cancer and ionizing radiation exposure are associated with inflammation. To identify a set of radiation-specific signatures of inflammation-associated genes in the blood of partially exposed radiotherapy patients, differential expression of 249 inflammatory genes was analyzed in blood samples from cancer patients and healthy individuals. The gene expression analysis on a cohort of 63 cancer patients (endometrial, head and neck, and prostate cancer) before and during radiotherapy (24 h, 48 h, ~1 week, ~4-8 weeks, and 1 month after the last fraction) identified 31 genes and 15 up- and 16 down-regulated genes. Transcription variability under normal conditions was determined using blood drawn on three separate occasions from four healthy donors. No difference in inflammatory expression between healthy donors and cancer patients could be detected prior to radiotherapy. Remarkably, repeated sampling of healthy donors revealed an individual endogenous inflammatory signature. Next, the potential confounding effect of concomitant inflammation was studied in the blood of seven healthy donors taken before and 24 h after a flu vaccine or ex vivo LPS (lipopolysaccharide) treatment; flu vaccination was not detected at the transcriptional level and LPS did not have any effect on the radiation-induced signature identified. Finally, we identified a radiation-specific signature of 31 genes in the blood of radiotherapy patients that were common for all cancers, regardless of the immune status of patients. Confirmation via MQRT-PCR was obtained for BCL6, MYD88, MYC, IL7, CCR4 and CCR7. This study offers the foundation for future research on biomarkers of radiation exposure, radiation sensitivity, and radiation toxicity for personalized radiotherapy treatment.

The International Association for the Study of Lung Cancer Early Lung Imaging Confederation Open-Source Deep Learning and Quantitative Measurement Initiative.

INTRODUCTION: With global adoption of computed tomography (CT) lung cancer screening, there is increasing interest to use artificial intelligence (AI) deep learning methods to improve the clinical management process. To enable AI research using an open-source, cloud-based, globally distributed, screening CT imaging data set and computational environment that are compliant with the most stringent international privacy regulations that also protect the intellectual properties of researchers, the International Association for the Study of Lung Cancer sponsored development of the Early Lung Imaging Confederation (ELIC) resource in 2018. The objective of this report is to describe the updated capabilities of ELIC and illustrate how this resource can be used for clinically relevant AI research. METHODS: In this second phase of the initiative, metadata and screening CT scans from two time points were collected from 100 screening participants in seven countries. An automated deep learning AI lung segmentation algorithm, automated quantitative emphysema metrics, and a quantitative lung nodule volume measurement algorithm were run on these scans. RESULTS: A total of 1394 CTs were collected from 697 participants. The LAV950 quantitative emphysema metric was found to be potentially useful in distinguishing lung cancer from benign cases using a combined slice thickness more than or equal to 2.5 mm. Lung nodule volume change measurements had better sensitivity and specificity for classifying malignant from benign lung nodules when applied to solid lung nodules from high-quality CT scans. CONCLUSIONS: These initial experiments revealed that ELIC can support deep learning AI and quantitative imaging analyses on diverse and globally distributed cloud-based data sets.

miR-122 and miR-21 are Stable Components of miRNA Signatures of Early Lung Cancer after Validation in Three Independent Cohorts.

Several panels of circulating miRNAs have been reported as potential biomarkers of early lung cancer, yet the overlap of components between different panels is limited, and the universality of proposed biomarkers has been minimal across proposed panels. To assess the stability of the diagnostic potential of plasma miRNA signature of early lung cancer among different cohorts, a panel of 24 miRNAs tested in the frame of one lung cancer screening study (MOLTEST-2013, Poland) was validated with material collected in the frame of two other screening studies (MOLTEST-BIS, Poland; and SMAC, Italy) using the same standardized analytical platform (the miRCURY LNA miRNA PCR assay). On analysis of selected miRNAs, two associated with lung cancer development, miR-122 and miR-21, repetitively differentiated healthy participants from individuals with lung cancer. Additionally, miR-144 differentiated controls from cases specifically in subcohorts with adenocarcinoma. Other tested miRNAs did not overlap in the three cohorts. Classification models based on neither a single miRNA nor multicomponent miRNA panels (24-mer and 7-mer) showed classification performance sufficient for a standalone diagnostic biomarker (AUC, 75%, 71%, and 53% in MOLTEST-2013, SMAC, and MOLTEST-BIS, respectively, in the 7-mer model). The performance of classification in the MOLTEST-BIS cohort with the lowest contribution of adenocarcinomas was increased when only this cancer type was considered (AUC, 60% in 7-mer model).

Body mass index and partial remission in 119 children with type 1 diabetes-a 6-year observational study.

Background/objective: This long-term study aimed to analyze the associations between BMI Z-score, HbA1c, and daily insulin requirement (DIR) and the prevalence and duration of partial remission (PR) in children and adolescents with type 1 diabetes (T1D). Methods: After retrieving retrospective data for 195 patients from their health records at 24, 48, and 72 months after T1D diagnosis, the study group was comprised of 119 (57 girls) children with a complete dataset for all 6 years. PR was defined according to the ISPAD guidelines. Analyses were carried out in the whole group and subgroups according to PR duration: no PR at all (NPR), PR lasting less than 2 years (PR < 2), and PR at least 2 years (PR ≥ 2). Results: PR was observed in 63% of the patients (78.9% of overweight and 100% of obese patients). NPR patients showed the lowest mean initial BMI Z-score [-0.65 ± 1.29 vs. 0.02 ± 1.42, (PR < 2), p = 0.01 and vs. 0.64 ± 1.43 (PR ≥ 2), p = 0.17]. The dissimilarity in BMI across patients declined over time. Within the NPR group, the initial mean BMI Z-score significantly increased within the first 2 years (unadjusted p < 0.001) and remained constant afterward. In the PR <2 group, the highest increase in BMI Z-score occurred after 4 years (p < 0.001) and then decreased (p = 0.04). In the PR ≥2, the BMI Z-score slightly decreased within the first 2 years (p = 0.02), then increased (p = 0.03) and remained unchanged for the last 2 years. Six years after T1D started, the mean DIRs do not differ among the patient groups (ANOVA p = 0.272). Conclusion: During 6 years of follow-up, PR occurred in almost two-thirds of the studied children including almost all overweight and obese children. We observed a gradual normalization of the BMI Z-score at the end of the follow-up. BMI Z-score increased slightly in children with no remission initially but remained later constant until the end of observation. In both remitter groups, the increase in BMI Z-score appeared later when the protective honeymoon period ended. Regardless of BMI Z-score, the ß-cell destruction process progresses, and after 6 years, the DIR is similar for all patients.

MiniMed 780GTM in children with type 1 diabetes under seven years of age: Prospective open-label, single-arm, double-center, follow-up study.

BACKGROUND: Given the steadily rising incidence of type 1 diabetes (T1D), particularly among the youngest preschool children, coupled with well-documented challenges of achieving and maintaining optimal metabolic control in this age group, there is a growing need for advanced technological devices. OBJECTIVE: To evaluate glycaemic control in children below the age of seven with type 1 diabetes (T1D) and assess the safety of the advanced hybrid closed loop (AHCL) system in comparison to the previous treatment method, a sensor-augmented pump with predictive low-glucose suspend (SAP-PLGS). METHOD: Data from 10 children (aged 2.60-6.98 years) with T1D who transitioned to the AHCL system from SAP-PLGS were analysed. SAP-PLGS records from two weeks prior to the initiation of AHCL were compared with records from the initial four weeks post-switch (excluding the training period). These data were examined at two 2-week intervals and compared with records from two weeks post six-month usage of the AHCL. RESULTS: A significant decrease in the average nighttime glucose concentration was observed compared to pre-AHCL values (p= 0.001, concordance W = 0.53). The Glucose Management Indicator (GMI) value significantly decreased from 6.88 ± 0.37% to 6.52 ± 0.32% (p= 0.018, rbc = 0.93) immediately following the device switch and stabilized at 6.50 ± 0.28% (p= 0.001, W = 0.53) and 6.55 ± 0.41% (p= 0.001, W = 0.53) at subsequent stages of the study. An improvement was also observed in mean glucose values for time spent < 54 mg/dl, while the proportion of time within this range was maintained, both during the day (p< 0.001, W = 0.58) and at night (p= 0.002, W = 0.83). CONCLUSION: The AHCL MiniMed 780GTM system improved glycaemic control in the studied group of children under seven years of age with T1D compared to previous SAP-PLGS therapy. It proved to be safe for delivering insulin in this age group.

Evaluation of zero counts to better understand the discrepancies between bulk and single-cell RNA-Seq platforms.

Recent advances in sample preparation and sequencing technology have made it possible to profile the transcriptomes of individual cells using single-cell RNA sequencing (scRNA-Seq). Compared to bulk RNA-Seq data, single-cell data often contain a higher percentage of zero reads, mainly due to lower sequencing depth per cell, which affects mostly measurements of low-expression genes. However, discrepancies between platforms are observed regardless of expression level. Using four paired datasets with multiple samples each, we investigated technical and biological factors that can contribute to this expression shift. Using two separate machine learning models we found that, in addition to expression level, RNA integrity, gene or UTR3 length, and the number of transcripts potentially also influence the occurrence of zeros. These findings could enable the development of novel analytical methods for cross-platform expression shift correction. We also identified genes and biological pathways in our diverse datasets that consistently showed differences when assessed at the single cell versus bulk level to assist in interpreting analysis across transcriptomic platforms. At the gene level, 25 genes (0.12%) were found in all datasets as discordant, but at the pathway level, 7 pathways (2.02%) showed shared enrichment in discordant genes.

Multivariate piecewise linear regression model to predict radiosensitivity using the association with the genome-wide copy number variation.

Introduction: The search for biomarkers to predict radiosensitivity is important not only to individualize radiotherapy of cancer patients but also to forecast radiation exposure risks. The aim of this study was to devise a machine-learning method to stratify radiosensitivity and to investigate its association with genome-wide copy number variations (CNVs) as markers of sensitivity to ionizing radiation. Methods: We used the Affymetrix CytoScan HD microarrays to survey common CNVs in 129 fibroblast cell strains. Radiosensitivity was measured by the surviving fraction at 2 Gy (SF2). We applied a dynamic programming (DP) algorithm to create a piecewise (segmented) multivariate linear regression model predicting SF2 and to identify SF2 segment-related distinctive CNVs. Results: SF2 ranged between 0.1384 and 0.4860 (mean=0.3273 The DP algorithm provided optimal segmentation by defining batches of radio-sensitive (RS), normally-sensitive (NS), and radio-resistant (RR) responders. The weighted mean relative errors (MRE) decreased with increasing the segments' number. The borders of the utmost segments have stabilized after partitioning SF2 into 5 subranges. Discussion: The 5-segment model associated C-3SFBP marker with the most-RS and C-7IUVU marker with the most-RR cell strains. Both markers were mapped to gene regions (MCC and SLC1A6, respectively). In addition, C-3SFBP marker is also located in enhancer and multiple binding motifs. Moreover, for most CNVs significantly correlated with SF2, the radiosensitivity increased with the copy-number decrease.In conclusion, the DP-based piecewise multivariate linear regression method helps narrow the set of CNV markers from the whole radiosensitivity range to the smaller intervals of interest. Notably, SF2 partitioning not only improves the SF2 estimation but also provides distinctive markers. Ultimately, segment-related markers can be used, potentially with tissues' specific factors or other clinical data, to identify radiotherapy patients who are most RS and require reduced doses to avoid complications and the most RR eligible for dose escalation to improve outcomes.

Sex- and age-specific aspects of human peripheral T-cell dynamics.

Background: The diversity of the antigenic T cell receptor (TCR) repertoire clonally expressed on T lymphocytes is a key element of the adaptive immune system protective functions. A decline in diversity in the older adults is associated with health deterioration. This diversity is generated by the rearrangement of TRB genes coding for TCR chains during lymphocyte differentiation in the thymus, but is essentially maintained by peripheral T lymphocytes proliferation for most of life. Deep sequencing of rearranged TRB genes from blood cells allows the monitoring of peripheral T cell repertoire dynamics. We analysed two aspects of rearranged TRB diversity, related to T lymphocyte proliferation and to the distribution of the T cell clone size, in a collection of repertoires obtained from 1 to 74 years-old donors. Results: Our results show that peripheral T lymphocytes expansion differs according to the recombination status of their TRB loci. Their proliferation rate changes with age, with different patterns in men and women. T cell clone size becomes more heterogeneous with time, and, in adults, is always more even in women. Importantly, a longitudinal analysis of TRB repertoires obtained at ten years intervals from individual men and women confirms the findings of this cross-sectional study. Conclusions: Peripheral T lymphocyte proliferation partially depends on their thymic developmental history. The rate of proliferation of T cells differing in their TRB rearrangement status is different in men and women before the age of 18 years old, but similar thereafter.

GMM-Based Expanded Feature Space as a Way to Extract Useful Information for Rare Cell Subtypes Identification in Single-Cell Mass Cytometry.

Cell subtype identification from mass cytometry data presents a persisting challenge, particularly when dealing with millions of cells. Current solutions are consistently under development, however, their accuracy and sensitivity remain limited, particularly in rare cell-type detection due to frequent downsampling. Additionally, they often lack the capability to analyze large data sets. To overcome these limitations, a new method was suggested to define an extended feature space. When combined with the robust clustering algorithm for big data, it results in more efficient cell clustering. Each marker's intensity distribution is presented as a mixture of normal distributions (Gaussian Mixture Model, GMM), and the expanded space is created by spanning over all obtained GMM components. The projection of the initial flow cytometry marker domain into the expanded space employs GMM-based membership functions. An evaluation conducted on three established cellular identification algorithms (FlowSOM, ClusterX, and PARC) utilizing the most substantial publicly available annotated dataset by Samusik et al. demonstrated the superior performance of the suggested approach in comparison to the standard. Although our approach identified 20 cell clusters instead of the expected 24, their intra-cluster homogeneity and inter-cluster differences were superior to the 24-cluster FlowSOM-based solution.

Proteomic Profile Distinguishes New Subpopulations of Breast Cancer Patients with Different Survival Outcomes.

As a highly heterogeneous disease, breast cancer (BRCA) demonstrates a diverse molecular portrait. The well-established molecular classification (PAM50) relies on gene expression profiling. It insufficiently explains the observed clinical and histopathological diversity of BRCAs. This study aims to demographically and clinically characterize the six BRCA subpopulations (basal, HER2-enriched, and four luminal ones) revealed by their proteomic portraits. GMM-based high variate protein selection combined with PCA/UMAP was used for dimensionality reduction, while the k-means algorithm allowed patient clustering. The statistical analysis (log-rank and Gehan-Wilcoxon tests, hazard ratio HR as the effect size ES) showed significant differences across identified subpopulations in Disease-Specific Survival (p = 0.0160) and Progression-Free Interval (p = 0.0264). Luminal subpopulations vary in prognosis (Disease-Free Interval, p = 0.0277). The A2 subpopulation is of the poorest, comparable to the HER2-enriched subpopulation, prognoses (HR = 1.748, referenced to Luminal B, small ES), while A3 is of the best (HR = 0.250, large ES). Similar to PAM50 subtypes, no substantial dependency on demographic and clinical factors was detected across Luminal subpopulations, as measured by χ2 test and Cramér's V for ES, and ANOVA with appropriate post hocs combined with η2 or Cohen's d-type ES, respectively. Progesterone receptors can serve as the potential A2 biomarker within Luminal patients. Further investigation of molecular differences is required to examine the potential prognostic or clinical applications.

POLCOVID: a multicenter multiclass chest X-ray database (Poland, 2020-2021).

The outbreak of the SARS-CoV-2 pandemic has put healthcare systems worldwide to their limits, resulting in increased waiting time for diagnosis and required medical assistance. With chest radiographs (CXR) being one of the most common COVID-19 diagnosis methods, many artificial intelligence tools for image-based COVID-19 detection have been developed, often trained on a small number of images from COVID-19-positive patients. Thus, the need for high-quality and well-annotated CXR image databases increased. This paper introduces POLCOVID dataset, containing chest X-ray (CXR) images of patients with COVID-19 or other-type pneumonia, and healthy individuals gathered from 15 Polish hospitals. The original radiographs are accompanied by the preprocessed images limited to the lung area and the corresponding lung masks obtained with the segmentation model. Moreover, the manually created lung masks are provided for a part of POLCOVID dataset and the other four publicly available CXR image collections. POLCOVID dataset can help in pneumonia or COVID-19 diagnosis, while the set of matched images and lung masks may serve for the development of lung segmentation solutions.

Pathological changes or technical artefacts? The problem of the heterogenous databases in COVID-19 CXR image analysis.

BACKGROUND: When the COVID-19 pandemic commenced in 2020, scientists assisted medical specialists with diagnostic algorithm development. One scientific research area related to COVID-19 diagnosis was medical imaging and its potential to support molecular tests. Unfortunately, several systems reported high accuracy in development but did not fare well in clinical application. The reason was poor generalization, a long-standing issue in AI development. Researchers found many causes of this issue and decided to refer to them as confounders, meaning a set of artefacts and methodological errors associated with the method. We aim to contribute to this steed by highlighting an undiscussed confounder related to image resolution. METHODS: 20 216 chest X-ray images (CXR) from worldwide centres were analyzed. The CXRs were bijectively projected into the 2D domain by performing Uniform Manifold Approximation and Projection (UMAP) embedding on the radiomic features (rUMAP) or CNN-based neural features (nUMAP) from the pre-last layer of the pre-trained classification neural network. Additional 44 339 thorax CXRs were used for validation. The comprehensive analysis of the multimodality of the density distribution in rUMAP/nUMAP domains and its relation to the original image properties was used to identify the main confounders. RESULTS: nUMAP revealed a hidden bias of neural networks towards the image resolution, which the regular up-sampling procedure cannot compensate for. The issue appears regardless of the network architecture and is not observed in a high-resolution dataset. The impact of the resolution heterogeneity can be partially diminished by applying advanced deep-learning-based super-resolution networks. CONCLUSIONS: rUMAP and nUMAP are great tools for image homogeneity analysis and bias discovery, as demonstrated by applying them to COVID-19 image data. Nonetheless, nUMAP could be applied to any type of data for which a deep neural network could be constructed. Advanced image super-resolution solutions are needed to reduce the impact of the resolution diversity on the classification network decision.

Can T1-Weighted Magnetic Resonance Imaging Significantly Improve Mini-Mental State Examination-Based Distinguishing Between Mild Cognitive Impairment and Early-Stage Alzheimer's Disease?

BACKGROUND: Detecting early-stage Alzheimer's disease (AD) is still problematic in clinical practice. This work aimed to find T1-weighted MRI-based markers for AD and mild cognitive impairment (MCI) to improve the screening process. OBJECTIVE: Our assumption was to build a screening model that would be accessible and easy to use for physicians in their daily clinical routine. METHODS: The multinomial logistic regression was used to detect status: AD, MCI, and normal control (NC) combined with the Bayesian information criterion for model selection. Several T1-weighted MRI-based radiomic features were considered explanatory variables in the prediction model. RESULTS: The best radiomic predictor was the relative brain volume. The proposed method confirmed its quality by achieving a balanced accuracy of 95.18%, AUC of 93.25%, NPV of 97.93%, and PPV of 90.48% for classifying AD versus NC for the European DTI Study on Dementia (EDSD). The comparison of the two models: with the MMSE score only as an independent variable and corrected for the relative brain value and age, shows that the addition of the T1-weighted MRI-based biomarker improves the quality of MCI detection (AUC: 67.04% versus 71.08%) while maintaining quality for AD (AUC: 93.35% versus 93.25%). Additionally, among MCI patients predicted as AD inconsistently with the original diagnosis, 60% from ADNI and 76.47% from EDSD were re-diagnosed as AD within a 48-month follow-up. It shows that our model can detect AD patients a few years earlier than a standard medical diagnosis. CONCLUSION: The created method is non-invasive, inexpensive, clinically accessible, and efficiently supports AD/MCI screening.

Short- and long-term effects of radiation exposure at low dose and low dose rate in normal human VH10 fibroblasts.

Introduction: Experimental studies complement epidemiological data on the biological effects of low doses and dose rates of ionizing radiation and help in determining the dose and dose rate effectiveness factor. Methods: Human VH10 skin fibroblasts exposed to 25, 50, and 100 mGy of 137Cs gamma radiation at 1.6, 8, 12 mGy/h, and at a high dose rate of 23.4 Gy/h, were analyzed for radiation-induced short- and long-term effects. Two sample cohorts, i.e., discovery (n = 30) and validation (n = 12), were subjected to RNA sequencing. The pool of the results from those six experiments with shared conditions (1.6 mGy/h; 24 h), together with an earlier time point (0 h), constituted a third cohort (n = 12). Results: The 100 mGy-exposed cells at all abovementioned dose rates, harvested at 0/24 h and 21 days after exposure, showed no strong gene expression changes. DMXL2, involved in the regulation of the NOTCH signaling pathway, presented a consistent upregulation among both the discovery and validation cohorts, and was validated by qPCR. Gene set enrichment analysis revealed that the NOTCH pathway was upregulated in the pooled cohort (p = 0.76, normalized enrichment score (NES) = 0.86). Apart from upregulated apical junction and downregulated DNA repair, few pathways were consistently changed across exposed cohorts. Concurringly, cell viability assays, performed 1, 3, and 6 days post irradiation, and colony forming assay, seeded just after exposure, did not reveal any statistically significant early effects on cell growth or survival patterns. Tendencies of increased viability (day 6) and reduced colony size (day 21) were observed at 12 mGy/h and 23.4 Gy/min. Furthermore, no long-term changes were observed in cell growth curves generated up to 70 days after exposure. Discussion: In conclusion, low doses of gamma radiation given at low dose rates had no strong cytotoxic effects on radioresistant VH10 cells.

Combining Low-Dose Computer-Tomography-Based Radiomics and Serum Metabolomics for Diagnosis of Malignant Nodules in Participants of Lung Cancer Screening Studies.

Radiomics is an emerging approach to support the diagnosis of pulmonary nodules detected via low-dose computed tomography lung cancer screening. Serum metabolome is a promising source of auxiliary biomarkers that could help enhance the precision of lung cancer diagnosis in CT-based screening. Thus, we aimed to verify whether the combination of these two techniques, which provides local/morphological and systemic/molecular features of disease at the same time, increases the performance of lung cancer classification models. The collected cohort consists of 1086 patients with radiomic and 246 patients with serum metabolomic evaluations. Different machine learning techniques, i.e., random forest and logistic regression were applied for each omics. Next, model predictions were combined with various integration methods to create a final model. The best single omics models were characterized by an AUC of 83% in radiomics and 60% in serum metabolomics. The model integration only slightly increased the performance of the combined model (AUC equal to 85%), which was not statistically significant. We concluded that radiomics itself has a good ability to discriminate lung cancer from benign lesions. However, additional research is needed to test whether its combination with other molecular assessments would further improve the diagnosis of screening-detected lung nodules.

DiviK: divisive intelligent K-means for hands-free unsupervised clustering in big biological data.

BACKGROUND: Investigating molecular heterogeneity provides insights into tumour origin and metabolomics. The increasing amount of data gathered makes manual analyses infeasible-therefore, automated unsupervised learning approaches are utilised for discovering tissue heterogeneity. However, automated analyses require experience setting the algorithms' hyperparameters and expert knowledge about the analysed biological processes. Moreover, feature engineering is needed to obtain valuable results because of the numerous features measured. RESULTS: We propose DiviK: a scalable stepwise algorithm with local data-driven feature space adaptation for segmenting high-dimensional datasets. The algorithm is compared to the optional solutions (regular k-means, spatial and spectral approaches) combined with different feature engineering techniques (None, PCA, EXIMS, UMAP, Neural Ions). Three quality indices: Dice Index, Rand Index and EXIMS score, focusing on the overall composition of the clustering, coverage of the tumour region and spatial cluster consistency, are used to assess the quality of unsupervised analyses. Algorithms were validated on mass spectrometry imaging (MSI) datasets-2D human cancer tissue samples and 3D mouse kidney images. DiviK algorithm performed the best among the four clustering algorithms compared (overall quality score 1.24, 0.58 and 162 for d(0, 0, 0), d(1, 1, 1) and the sum of ranks, respectively), with spectral clustering being mostly second. Feature engineering techniques impact the overall clustering results less than the algorithms themselves (partial [Formula: see text] effect size: 0.141 versus 0.345, Kendall's concordance index: 0.424 versus 0.138 for d(0, 0, 0)). CONCLUSIONS: DiviK could be the default choice in the exploration of MSI data. Thanks to its unique, GMM-based local optimisation of the feature space and deglomerative schema, DiviK results do not strongly depend on the feature engineering technique applied and can reveal the hidden structure in a tissue sample. Additionally, DiviK shows high scalability, and it can process at once the big omics data with more than 1.5 mln instances and a few thousand features. Finally, due to its simplicity, DiviK is easily generalisable to an even more flexible framework. Therefore, it is helpful for other -omics data (as single cell spatial transcriptomic) or tabular data in general (including medical images after appropriate embedding). A generic implementation is freely available under Apache 2.0 license at https://github.com/gmrukwa/divik .

Influence of single-cell RNA sequencing data integration on the performance of differential gene expression analysis.

Large-scale comprehensive single-cell experiments are often resource-intensive and require the involvement of many laboratories and/or taking measurements at various times. This inevitably leads to batch effects, and systematic variations in the data that might occur due to different technology platforms, reagent lots, or handling personnel. Such technical differences confound biological variations of interest and need to be corrected during the data integration process. Data integration is a challenging task due to the overlapping of biological and technical factors, which makes it difficult to distinguish their individual contribution to the overall observed effect. Moreover, the choice of integration method may impact the downstream analyses, including searching for differentially expressed genes. From the existing data integration methods, we selected only those that return the full expression matrix. We evaluated six methods in terms of their influence on the performance of differential gene expression analysis in two single-cell datasets with the same biological study design that differ only in the way the measurement was done: one dataset manifests strong batch effects due to the measurements of each sample at a different time. Integrated data were visualized using the UMAP method. The evaluation was done both on individual gene level using parametric and non-parametric approaches for finding differentially expressed genes and on gene set level using gene set enrichment analysis. As an evaluation metric, we used two correlation coefficients, Pearson and Spearman, of the obtained test statistics between reference, test, and corrected studies. Visual comparison of UMAP plots highlighted ComBat-seq, limma, and MNN, which reduced batch effects and preserved differences between biological conditions. Most of the tested methods changed the data distribution after integration, which negatively impacts the use of parametric methods for the analysis. Two algorithms, MNN and Scanorama, gave very poor results in terms of differential analysis on gene and gene set levels. Finally, we highlight ComBat-seq as it led to the highest correlation of test statistics between reference and corrected dataset among others. Moreover, it does not distort the original distribution of gene expression data, so it can be used in all types of downstream analyses.

CMB-HUNT: Automatic detection of cerebral microbleeds using a deep neural network.

Cerebral microbleeds (CMBs) are gaining increasing interest due to their importance in diagnosing cerebral small vessel diseases. However, manual inspection of CMBs is time-consuming and prone to human error. Existing automated or semi-automated solutions still have insufficient detection sensitivity and specificity. Furthermore, they frequently use more than one magnetic resonance imaging modality, but these are not always available. The majority of AI-based solutions use either numeric or image data, which may not provide sufficient information about the true nature of CMBs. This paper proposes a deep neural network with multi-type input data for automated CMB detection (CMB-HUNT) using only susceptibility-weighted imaging data (SWI). Combination of SWIs and radiomic-type numerical features allowed us to identify CMBs with high accuracy without the need for additional imaging modalities or complex predictive models. Two independent datasets were used: one with 304 patients (39 with CMBs) for training and internal system validation and one with 61 patients (21 with CMBs) for external validation. For the hold-out testing dataset, CMB-HUNT reached a sensitivity of 90.0%. As results of testing showed, CMB-HUNT outperforms existing methods in terms of the number of FPs per case, which is the lowest reported thus far (0.54 FPs/patient). The proposed system was successfully applied to the independent validation set, reaching a sensitivity of 91.5% with 1.9 false positives per patient and proving its generalization potential. The results were comparable to previous studies. Our research confirms the usefulness of deep learning solutions for CMB detection based only on one MRI modality.