Unsupervised Clustering Approach Research Articles

Exploring the butyrate metabolism-related shared genes in metabolic associated steatohepatitis and ulcerative colitis

Metabolic-associated steatohepatitis (MASH) and ulcerative colitis (UC) exhibit a complex interconnection with immune dysfunction, dysbiosis of the gut microbiota, and activation of inflammatory pathways. This study aims to identify and validate critical butyrate metabolism-related shared genes between both UC and MASH. Clinical information and gene expression profiles were sourced from the Gene Expression Omnibus (GEO) database. Shared butyrate metabolism-related differentially expressed genes (sBM-DEGs) between UC and MASH were identified via various bioinformatics methods. Functional enrichment analysis was performed, and UC patients were categorized into subtypes using the consensus clustering algorithm based on sBM-DEGs. Key genes within sBM-DEGs were screened through Random Forest, Support Vector Machines-Recursive Feature Elimination, and Light Gradient Boosting. The diagnostic efficacy of these genes was evaluated using receiver operating characteristic (ROC) analysis on independent datasets. Additionally, the expression levels of characteristic genes were validated across multiple independent datasets and human specimens. Forty-nine shared DEGs between UC and MASH were identified, with enrichment analysis highlighting significant involvement in immune, inflammatory, and metabolic pathways. The intersection of butyrate metabolism-related genes with these DEGs produced 10 sBM-DEGs. These genes facilitated the identification of molecular subtypes of UC patients using an unsupervised clustering approach. ANXA5, CD44, and SLC16A1 were pinpointed as hub genes through machine learning algorithms and feature importance rankings. ROC analysis confirmed their diagnostic efficacy in UC and MASH across various datasets. Additionally, the expression levels of these three hub genes showed significant correlations with immune cells. These findings were validated across independent datasets and human specimens, corroborating the bioinformatics analysis results. Integrated bioinformatics identified three significant biomarkers, ANXA5, CD44, and SLC16A1, as DEGs linked to butyrate metabolism. These findings offer new insights into the role of butyrate metabolism in the pathogenesis of UC and MASH, suggesting its potential as a valuable diagnostic biomarker.

Cluster analysis of cancer knowledge, attitudes and behaviors in the Moroccan population

BackgroundCancer has become a major health concern due to the increasing morbidity and mortality rates, and its negative social, economic consequences and the heavy financial burden incurred by cancer patients. About 40% of cancers are preventable. The aim of this study was to assess the knowledge, attitudes, and practices regarding cancer prevention, and associated characteristics to inform the development of targeted cancer prevention campaigns and policies.MethodsWe conducted a cross-sectional survey of adult patients at Mohamed Sekkat and Sidi Othmane Hospitals in Casablanca, Morocco. Data collection was conducted by two trained interviewers who administered the questionnaire in-person in the local language. An unsupervised clustering approach included 17 candidate variables for the cluster analysis. The variables covered a wide range of characteristics, including demographics, health perceptions and attitudes. Survey answers were calculated to compose qualitative ordinal categories, including a cancer attitude score and knowledge score.ResultsThe cluster-based analysis showed that participants in cluster 1 had the highest mean attitude score (13.9 ± 2.15) and percentage of individuals with a high level of knowledge (50.8%) whereas the lowest mean attitude score (9.48 ± 2.02) and knowledge level (7.5%.) were found in cluster 3. The participants with the lowest cancer attitude scores and knowledge levels were aged 34 to 47 years old (middle age group), predominantly females, living in rural settings, and were least likely to report health professionals as a source of health information.ConclusionsThe findings showed that female individuals living in rural settings, belonging to an older age group, who were least likely to use health professionals as an information source had the lowest levels of knowledge and attitudes. These groups are amenable to targeted and tailored interventions aiming to modify their understanding of cancer in order to enhance the outcomes of Morocco’s on-going efforts in cancer prevention and control strategies.

Proposing unsupervised clustering-based earthquake records selection framework for computationally efficient nonlinear response history analysis of structures equipped with multi-stage friction pendulum bearings

The design and analysis of base-isolated structures are critical for ensuring seismic performance and safety. Nevertheless, to effectively design and analyze base-isolated structures, it is essential to consider the structure's response under various ground motions that may occur during an earthquake. The nonlinear response history analysis (NRHA) is a common method for evaluating structures' seismic response under a specific set of ground motions. However, conducting this analysis for many earthquakes can be computationally expensive and time-consuming. Therefore, it is desirable to have a method for selecting a subset of earthquake records representative of the earthquake population. Accordingly, this study introduces a novel block-based framework based on the K-means unsupervised clustering techniques for efficient earthquake records' selection NRHA of base-isolated structures. The proposed approach aims to reduce the number of records required for investigations by identifying clusters of similar seismic signals and selecting representative records from each cluster for the NRHA. Additionally, the proposed framework serves a vital role in identifying the earthquake records that can be used as a training dataset for developing artificial neural network (ANN) models, enabling rapid response estimation of base-isolated structures. This dual functionality of the proposed method provides significant value for researchers and engineers working in earthquake-prone countries. This article presents two case studies to demonstrate the applicability and effectiveness of the proposed framework. The first case study focuses on selecting a subset of earthquake records from a suit that meets the ASCE 7–22 requirements for NRHA on base-isolated structures, aiming to achieve the close mean responses between the full suit and the subset of records. The second case study highlights the development of an ANN model for response estimation of such structures using the proposed framework to select suitable earthquakes as the training dataset, hence providing an adequate database to achieve high estimation results. In general, the study results highlight the capabilities of the proposed frameworks and the benefits of the K-means unsupervised clustering approach in addressing the gap in the literature and providing a robust method for earthquake record selection in various analysis scenarios.

Identification of inflammatory clusters in long-COVID through analysis of plasma biomarker levels.

Mechanisms underlying long COVID remain poorly understood. Patterns of immunological responses in individuals with long COVID may provide insight into clinical phenotypes. Here we aimed to identify these immunological patterns and study the inflammatory processes ongoing in individuals with long COVID. We applied an unsupervised hierarchical clustering approach to analyze plasma levels of 42 biomarkers measured in individuals with long COVID. Logistic regression models were used to explore associations between biomarker clusters, clinical variables, and symptom phenotypes. In 101 individuals, we identified three inflammatory clusters: a limited immune activation cluster, an innate immune activation cluster, and a systemic immune activation cluster. Membership in these inflammatory clusters did not correlate with individual symptoms or symptom phenotypes, but was associated with clinical variables including age, BMI, and vaccination status. Differences in serologic responses between clusters were also observed. Our results indicate that clinical variables of individuals with long COVID are associated with their inflammatory profiles and can provide insight into the ongoing immune responses.

Identifying Bladder Phenotypes After Spinal Cord Injury With Unsupervised Machine Learning: A New Way to Examine Urinary Symptoms and Quality of Life.

Patients with spinal cord injuries (SCIs) experience variable urinary symptoms and quality of life (QOL). Our objective was to use machine learning to identify bladder-relevant phenotypes after SCI and assess their association with urinary symptoms and QOL. We used data from the Neurogenic Bladder Research Group SCI registry. Baseline variables that were previously shown to be associated with bladder symptoms/QOL were included in the machine learning environment. An unsupervised consensus clustering approach (k-prototypes) was used to identify 4 patient clusters. After qualitative review of the clusters, 2 outcomes of interest were assessed: the total Neurogenic Bladder Symptom Score (NBSS) and the NBSS-satisfaction question (QOL). The NBSS and NBSS-satisfaction question at baseline and after 1 year were compared between clusters using analysis of variance and linear regression. Among the 1263 included participants, the 4 identified clusters were termed "female predominant," "high function, low SCI complication," "quadriplegia with bowel/bladder morbidity," and "older, high SCI complication." Using outcome data from baseline, significant differences were observed in the NBSS score, with the female predominant group exhibiting worse bladder symptoms. After 1 year, the overall bladder symptoms (NBSS Total) did not change significantly by cluster; however, the QOL score for the high function, low SCI complication group had more improvement (β = -0.12, P = .005), while the female predominant group had more deterioration (β = 0.09, P = .047). This study demonstrates the utility of machine learning in uncovering bladder-relevant phenotypes among SCI patients. Future research should explore cluster-based targeted strategies to enhance bladder-related outcomes and QOL in SCI.

Limitations of clustering with PCA and correlated noise

It is now common to have a modest to large number of features on individuals with complex diseases. Unsupervised analyses, such as clustering with and without preprocessing by Principle Component Analysis (PCA), is widely used in practice to uncover subgroups in a sample. However, in many modern studies features are often highly correlated and noisy (e.g. SNP's, -omics, quantitative imaging markers, and electronic health record data). The practical performance of clustering approaches in these settings remains unclear. Through extensive simulations and empirical examples applying Gaussian Mixture Models and related clustering methods, we show these approaches (including variants of kmeans, VarSelLCM, HDClassifier, and Fisher-EM) can have very poor performance in many settings. We also show the poor performance is often driven by either an explicit or implicit assumption by the clustering algorithm that high variance features are relevant while lower variance features are irrelevant, called the variance as relevance assumption. We develop practical pre-processing approaches that improve analysis performance in some cases. This work offers practical guidance on the strengths and limitations of unsupervised clustering approaches in modern data analysis applications.

Using unsupervised clustering approaches to identify common mental health profiles and associated mental health-care service-use patterns in Ontario, Canada.

Mental health is a complex, multidimensional concept that goes beyond clinical diagnoses, including psychological distress, life stress, and well-being. In this study, we aimed to use unsupervised clustering approaches to identify multidimensional mental health profiles that exist in the population, and their associated service-use patterns. The data source was the 2012 Canadian Community Health Survey-Mental Health, linked to administrative health-care data; all Ontario, Canada, adult respondents were included. We used a partitioning around medoids clustering algorithm with Gower's proximity to identify groups with distinct combinations of mental health indicators and described them according to their sociodemographic and service-use characteristics. We identified 4 groups with distinct mental health profiles, including 1 group that met the clinical threshold for a depressive diagnosis, with the remaining 3 groups expressing differences in positive mental health, life stress, and self-rated mental health. The 4 groups had different age, employment, and income profiles and exhibited differential access to mental health-care services. This study represents the first step in identifying complex profiles of mental health at the population level in Ontario. Further research is required to better understand the potential causes and consequences of belonging to each of the mental health profiles identified. This article is part of a Special Collection on Mental Health.

Application of machine learning for identification of heterotic groups in sunflower through combined approach of phenotyping, genotyping and protein profiling

Application of machine learning in plant breeding is a recent concept, that has to be optimized for precise utilization in the breeding program of high yielding crop plants. Identification and efficient utilization of heterotic grouping pattern aided with machine learning approaches is of utmost importance in hybrid cultivar breeding as it can save time and resources required to breed a new plant hybrid/variety. In the present study, 109 genotypes of sunflower were investigated at morphological, biochemical (SDS-PAGE) and molecular levels (through micro-satellites (SSR) markers) for heterotic grouping. All the three datasets were combined, scaled, and subjected to unsupervised machine learning algorithms, i.e., Hierarchical clustering, K-means clustering and hybrid clustering algorithm (hierarchical + K-means) for assessment of efficiency and resolution power of these algorithms in practical plant breeding for heterotic grouping identification. Following the application of machine learning unsupervised clustering approach, two major groups were identified in the studied sunflower germplasm, and further classification revealed six smaller classes in each major group through hierarchical and hybrid clustering approach. Due to high resolution, obtained in hierarchical clustering, classification achieved through this algorithm was further used for selection of potential parents. One genotype from each smaller group was selected based on the maximum seed yield potential and hybridized in a line × tester mating design producing 36 F1 cross combinations. These F1s along with their parents were studied in open field conditions for validating the efficacy of identified heterotic groups in sunflowers genetic material under study. Data for 11 agronomic and qualitative traits were recorded. These 36 F1 combinations were tested for their combining ability (General/Specific), heterosis, genotypic and phenotypic correlation and path analysis. Results suggested that F1 hybrids performed better for all the traits under investigation than their respective parents. Findings of the study validated the use of machine learning approaches in practical plant breeding; however, more accurate and robust clustering algorithms need to be developed to handle the data noisiness of open field experiments.

Brain Tumor Detection and Categorization with Segmentation of Improved Unsupervised Clustering Approach and Machine Learning Classifier.

There is no doubt that brain tumors are one of the leading causes of death in the world. A biopsy is considered the most important procedure in cancer diagnosis, but it comes with drawbacks, including low sensitivity, risks during biopsy treatment, and a lengthy wait for results. Early identification provides patients with a better prognosis and reduces treatment costs. The conventional methods of identifying brain tumors are based on medical professional skills, so there is a possibility of human error. The labor-intensive nature of traditional approaches makes healthcare resources expensive. A variety of imaging methods are available to detect brain tumors, including magnetic resonance imaging (MRI) and computed tomography (CT). Medical imaging research is being advanced by computer-aided diagnostic processes that enable visualization. Using clustering, automatic tumor segmentation leads to accurate tumor detection that reduces risk and helps with effective treatment. This study proposed a better Fuzzy C-Means segmentation algorithm for MRI images. To reduce complexity, the most relevant shape, texture, and color features are selected. The improved Extreme Learning machine classifies the tumors with 98.56% accuracy, 99.14% precision, and 99.25% recall. The proposed classifier consistently demonstrates higher accuracy across all tumor classes compared to existing models. Specifically, the proposed model exhibits accuracy improvements ranging from 1.21% to 6.23% when compared to other models. This consistent enhancement in accuracy emphasizes the robust performance of the proposed classifier, suggesting its potential for more accurate and reliable brain tumor classification. The improved algorithm achieved accuracy, precision, and recall rates of 98.47%, 98.59%, and 98.74% on the Fig share dataset and 99.42%, 99.75%, and 99.28% on the Kaggle dataset, respectively, which surpasses competing algorithms, particularly in detecting glioma grades. The proposed algorithm shows an improvement in accuracy, of approximately 5.39%, in the Fig share dataset and of 6.22% in the Kaggle dataset when compared to existing models. Despite challenges, including artifacts and computational complexity, the study's commitment to refining the technique and addressing limitations positions the improved FCM model as a noteworthy advancement in the realm of precise and efficient brain tumor identification.

DNA-methylome-derived epigenetic fingerprint as an immunophenotype indicator of durable clinical immunotherapeutic benefits in head and neck squamous cell carcinoma.

Cancer immunotherapy provides durable response and improves survival in a subset of head and neck squamous cell carcinoma (HNSC) patients, which may due to discriminative tumor microenvironment (TME). Epigenetic regulations play critical roles in HNSC tumorigenesis, progression, and activation of functional immune cells. This study aims to identify an epigenetic signature as an immunophenotype indicator of durable clinical immunotherapeutic benefits in HNSC patients. Unsupervised consensus clustering approach was applied to distinguish immunophenotypes based on five immune signatures in The Cancer Genome Atlas (TCGA) HNSC cohort. Two immunophenotypes (immune 'Hot' and immune 'Cold') that had different TME features, diverse prognosis, and distinct DNA methylation patterns were recognized. Immunophenotype-related methylated signatures (IPMS) were identified by the least absolute shrinkage and selector operation algorithm. Additionally, the IPMS score by deconvolution algorithm was constructed as an immunophenotype classifier to predict clinical outcomes and immunotherapeutic response. The 'Hot' HNSC immunophenotype had higher immunoactivity and better overall survival (p = 0.00055) compared to the 'Cold' tumors. The immunophenotypes had distinct DNA methylation patterns, which was closely associated with HNSC tumorigenesis and functional immune cell infiltration. 311 immunophenotype-related methylated CpG sites (IRMCs) was identified from TCGA-HNSC dataset. IPMS score model achieved a strong clinical predictive performance for classifying immunophenotypes. The area under the curve value (AUC) of the IPMS score model reached 85.9% and 89.8% in TCGA train and test datasets, respectively, and robustness was verified in five HNSC validation datasets. It was also validated as an immunophenotype classifier for predicting durable clinical benefits (DCB) in lung cancer patients who received anti-PD-1/PD-L1 immunotherapy (p = 0.017) and TCGA-SKCM patients who received distinct immunotherapy (p = 0.033). This study systematically analyzed DNA methylation patterns in distinct immunophenotypes to identify IPMS with clinical prognostic potential for personalized epigenetic anticancer approaches in HNSC patients. The IPMS score model may serve as a reliable epigenome prognostic tool for clinical immunophenotyping to guide immunotherapeutic strategies in HNSC.

Abstract WP266: Prediction of Stroke Incidence Using Machine Learning: The Suita Study

Background: This population-based study investigated the potential of machine learning algorithms to predict stroke incidence and identify important risk factors. This study aimed to evaluate the accuracy of these algorithms in constructing a stroke prediction model. Methods: Participants from the Suita study were included, and baseline measurements were used to predict stroke outcomes over a 15-year follow-up period. In total, 7,389 participants and 51 variables were investigated, including demographics, medical history, medical imaging, laboratory data, and lifestyle habits. Initially, unsupervised K-prototype clustering was used to group participants based on their stroke risk. Subsequently, five supervised models (logistic regression, random forest, support vector machine, extreme gradient boosting, and light gradient boosted machine) were applied to predict the stroke outcomes. The Shapley Additive Explanations (SHAP) method determined the most critical variables. Results: Unsupervised clustering revealed significant differences in stroke incidence among the three identified risk clusters (9.1%, 6.6%, and 3.2%). These clusters were categorized into high-, medium-, and low-risk groups. Among the supervised models, the random forest algorithm demonstrated the best performance. The top ten most important variables for predicting stroke incidence were identified using the SHAP, with age being the most influential variable. Other significant risk markers included systolic blood pressure, hypertension, estimated glomerular filtration rate, metabolic syndrome, and blood sugar level. Additionally, elbow joint thickness and fructosamine, hemoglobin, and calcium levels were found to be potential predictors of stroke risk. Notably, the variables identified by the SHAP were consistent with those obtained from the unsupervised clustering approach in the high-risk group. Conclusion: Machine learning algorithms provide accurate predictions of stroke incidence and offer valuable insights into subclinical markers without the need for prior assumptions of causality. This study presents a data-driven machine-learning framework for stroke risk prediction and biomarker identification.

A machine learning approach for real-time cortical state estimation

Objective. Cortical function is under constant modulation by internally-driven, latent variables that regulate excitability, collectively known as ‘cortical state’. Despite a vast literature in this area, the estimation of cortical state remains relatively ad hoc, and not amenable to real-time implementation. Here, we implement robust, data-driven, and fast algorithms that address several technical challenges for online cortical state estimation. Approach. We use unsupervised Gaussian mixture models to identify discrete, emergent clusters in spontaneous local field potential signals in cortex. We then extend our approach to a temporally-informed hidden semi-Markov model (HSMM) with Gaussian observations to better model and infer cortical state transitions. Finally, we implement our HSMM cortical state inference algorithms in a real-time system, evaluating their performance in emulation experiments. Main results. Unsupervised clustering approaches reveal emergent state-like structure in spontaneous electrophysiological data that recapitulate arousal-related cortical states as indexed by behavioral indicators. HSMMs enable cortical state inferences in a real-time context by modeling the temporal dynamics of cortical state switching. Using HSMMs provides robustness to state estimates arising from noisy, sequential electrophysiological data. Significance. To our knowledge, this work represents the first implementation of a real-time software tool for continuously decoding cortical states with high temporal resolution (40 ms). The software tools that we provide can facilitate our understanding of how cortical states dynamically modulate cortical function on a moment-by-moment basis and provide a basis for state-aware brain machine interfaces across health and disease.

Mel-frequency cepstral coefficients outperform embeddings from pre-trained convolutional neural networks under noisy conditions for discrimination tasks of individual gibbons

Passive acoustic monitoring – an approach that utilizes autonomous acoustic recording units – allows for non-invasive monitoring of individuals, assuming that it is possible to acoustically distinguish individuals. However, identifying effective analytical approaches for individual identification remains a challenge. Our study investigates how the use of different feature representations impacts our ability to distinguish between individual female Northern grey gibbons (Hylobates funereus). We broadcast pre-recorded calls from twelve gibbon females and re-recorded the calls at varying distances (directly under the tree to ∼400 m away) using autonomous recording units. We evaluated the effectiveness of using different automated feature extraction approaches to classify gibbon calls: Mel-frequency cepstral coefficients (MFCCs), embeddings from three pre-trained neural networks (BirdNET, VGGish, and Wav2Vec2), and four commonly used acoustic indices. We used a supervised classification approach (random forest) to classify calls to the respective female and compared two unsupervised clustering approaches (affinity propagation clustering and hierarchical density-based spatial clustering) to evaluate which features were most effective for distinguishing female calls without using class labels. We used MFCCs as a baseline as previous work has shown they can be used to distinguish high-quality calls of individual gibbon females. Human annotators could only identify calls in spectrograms from recordings <350 m from the playback speaker with signal-to-noise ratio ∼ 0 dB, so our results focus on these recordings. Using supervised classification, our results confirmed the efficiency of MFCCs and the use of embeddings from one neural network (BirdNET) for effective acoustic classification of gibbon individuals at closer recording distances (signal-to-noise ratio > 10 dB), while the remaining features did not perform well. Contrary to our expectations, we found that MFCCs outperformed all other features for the unsupervised clustering tasks at closer distances and none of the features performed well at farther distances. The ability to acoustically discriminate animals under noisy conditions and from low signal-to-noise ratio calls has important implications for monitoring populations of endangered animals, such as gibbons. Focusing only on high signal-to-noise ratio calls for individual discrimination may not be possible for rare sounds, and future work should focus on developing effective approaches of feature extraction that can perform well across noisy, real-world conditions with a limited number of training samples.

New HCC Subtypes Based on CD8 Tex-Related lncRNA Signature Could Predict Prognosis, Immunological and Drug Sensitivity Characteristics of Hepatocellular Carcinoma.

Hepatocellular carcinoma has become one of the severe diseases threatening human health. T cell exhaustion is deemed as a reason for immunotherapy resistance. However, little is known about the roles of CD8 Tex-related lncRNAs in HCC. We processed single-cell RNA sequencing to identify CD8 Tex-related genes. CD8 Tex-related lncRNAs were identified based on their correlations with mRNAs. Unsupervised clustering approach was used to identify molecular clusters of CD8 Tex-related lncRNAs. Differences in prognosis and immune infiltration between the clusters were explored. Machine learning algorithms were used to construct a prognostic signature. Samples were classified as low- and high-risk groups based on their risk scores. We identified prognosis-related lncRNAs and constructed a ceRNA network. In vitro experiments were conducted to investigate the impacts of CD8 Tex-related lncRNAs on proliferation and apoptosis of HCC cells. We clarified cell types within two HCC single-cell datasets. We identified specific markers of CD8 Tex cells and analyzed their potential functions. Twenty-eight lncRNAs were identified as CD8 Tex-related. Based on CD8 Tex-related lncRNAs, samples were categorized into two distinct clusters, which exhibited significant differences in survival rates and immune infiltration. Ninety-six algorithm combinations were employed to establish a prognostic signature. RSF emerged as the one with the highest C-index. Patients in high- and low-risk groups exhibited marked differences in prognosis, enriched pathways, mutations and drug sensitivities. MCM3AP-AS1, MAPKAPK5-AS1 and PART1 were regarded as prognosis-related lncRNAs. A ceRNA network was constructed based on CD8 Tex-related lncRNAs and mRNAs. Experiments on cell lines and organoids indicated that downregulation of MCM3AP-AS1, MAPKAPK5-AS1 and PART1 suppressed cell proliferation and induced apoptosis. CD8 Tex-related lncRNAs played crucial roles in HCC progression. Our findings provided new insights into the regulatory mechanisms of CD8 Tex-related lncRNAs in HCC.

Unraveling Earthquake Clusters Composing the 2014 Alto Tiberina Earthquake Swarm via Unsupervised Learning

AbstractEarthquake swarms represent a particular mode of seismicity, not directly related to the occurrence of large earthquakes (e.g., aftershocks) but rather driven by external forcing such as aseismic deformation or fluid migration in fault systems. Sometimes their occurrence overlaps with observable geodetic signals in space and time, indicating a direct link. However, the low resolution of geodetic observations tends to obscure the small scale spatial and temporal dynamics of swarms. In this work, we automatically extract clusters of seismicity related to the 2014 Alto Tiberina swarm sequence (Italy) using an unsupervised clustering approach that exploits space and time information of the seismicity. The quantitative characterization of each cluster indicates that the overall swarm is composed of spatially and temporally confined (sub) swarms each of which could potentially be driven by small‐scale aseismic deformation process. This observation aligns with similar findings during slow slip events in subduction zones.

Glioblastoma pseudoprogression and true progression reveal spatially variable transcriptional differences

Post-resection radiologic monitoring to identify areas of new or progressive enhancement concerning for cancer recurrence is critical during patients with glioblastoma follow-up. However, treatment-related pseudoprogression presents with similar imaging features but requires different clinical management. While pathologic diagnosis is the gold standard to differentiate true progression and pseudoprogression, the lack of objective clinical standards and admixed histologic presentation creates the needs to (1) validate the accuracy of current approaches and (2) characterize differences between these entities to objectively differentiate true disease. We demonstrated using an online RNAseq repository of recurrent glioblastoma samples that cancer-immune cell activity levels correlate with heterogenous clinical outcomes in patients. Furthermore, nCounter RNA expression analysis of 48 clinical samples taken from second neurosurgical resection supports that pseudoprogression gene expression pathways are dominated with immune activation, whereas progression is predominated with cell cycle activity. Automated image processing and spatial expression analysis however highlight a failure to apply these broad expressional differences in a subset of cases with clinically challenging admixed histology. Encouragingly, applying unsupervised clustering approaches over our segmented histologic images provides novel understanding of morphologically derived differences between progression and pseudoprogression. Spatially derived data further highlighted polarization of myeloid populations that may underscore the tumorgenicity of novel lesions. These findings not only help provide further clarity of potential targets for pathologists to better assist stratification of progression and pseudoprogression, but also highlight the evolution of tumor-immune microenvironment changes which promote tumor recurrence.

Rarity: discovering rare cell populations from single-cell imaging data.

Cell type identification plays an important role in the analysis and interpretation of single-cell data and can be carried out via supervised or unsupervised clustering approaches. Supervised methods are best suited where we can list all cell types and their respective marker genes a priori, while unsupervised clustering algorithms look for groups of cells with similar expression properties. This property permits the identification of both known and unknown cell populations, making unsupervised methods suitable for discovery. Success is dependent on the relative strength of the expression signature of each group as well as the number of cells. Rare cell types therefore present a particular challenge that is magnified when they are defined by differentially expressing a small number of genes. Typical unsupervised approaches fail to identify such rare subpopulations, and these cells tend to be absorbed into more prevalent cell types. In order to balance these competing demands, we have developed a novel statistical framework for unsupervised clustering, named Rarity, that enables the discovery process for rare cell types to be more robust, consistent, and interpretable. We achieve this by devising a novel clustering method based on a Bayesian latent variable model in which we assign cells to inferred latent binary on/off expression profiles. This lets us achieve increased sensitivity to rare cell populations while also allowing us to control and interpret potential false positive discoveries. We systematically study the challenges associated with rare cell type identification and demonstrate the utility of Rarity on various IMC datasets. Implementation of Rarity together with examples is available from the Github repository (https://github.com/kasparmartens/rarity).

Unsupervised machine learning for risk stratification and identification of relevant subgroups of ascending aorta dimensions using cardiac CT and clinical data

The potential of precision population health lies in its capacity to utilize robust patient data for customized prevention and care targeted at specific groups. Machine learning has the potential to automatically identify clinically relevant subgroups of individuals, considering heterogeneous data sources. This study aimed to assess whether unsupervised machine learning (UML) techniques could interpret different clinical data to uncover clinically significant subgroups of patients suspected of coronary artery disease and identify different ranges of aorta dimensions in the different identified subgroups. We employed a random forest-based cluster analysis, utilizing 14 variables from 1170 (717 men/453 women) participants. The unsupervised clustering approach successfully identified four distinct subgroups of individuals with specific clinical characteristics, and this allows us to interpret and assess different ranges of aorta dimensions for each cluster. By employing flexible UML algorithms, we can effectively process heterogeneous patient data and gain deeper insights into clinical interpretation and risk assessment.

Ensemble classification of integrated CT scan datasets in detecting COVID-19 using feature fusion from contourlet transform and CNN

The COVID-19 disease caused by coronavirus is constantly changing due to the emergence of different variants and thousands of people are dying every day worldwide. Early detection of this new form of pulmonary disease can reduce the mortality rate. In this paper, an automated method based on machine learning (ML) and deep learning (DL) has been developed to detect COVID-19 using computed tomography (CT) scan images extracted from three publicly available datasets (A total of 11,407 images; 7397 COVID-19 images and 4010 normal images). An unsupervised clustering approach that is a modified region-based clustering technique for segmenting COVID-19 CT scan image has been proposed. Furthermore, contourlet transform and convolution neural network (CNN) have been employed to extract features individually from the segmented CT scan images and to fuse them in one feature vector. Binary differential evolution (BDE) approach has been employed as a feature optimization technique to obtain comprehensible features from the fused feature vector. Finally, a ML/DL-based ensemble classifier considering bagging technique has been employed to detect COVID-19 from the CT images. A fivefold and generalization cross-validation techniques have been used for the validation purpose. Classification experiments have also been conducted with several pre-trained models (AlexNet, ResNet50, GoogleNet, VGG16, VGG19) and found that the ensemble classifier technique with fused feature has provided state-of-the-art performance with an accuracy of 99.98%.

Leveraging transcriptomics to develop bronchopulmonary dysplasia endotypes: a concept paper

ImpactBronchopulmonary dysplasia has multiple definitions that are currently based on phenotypic characteristics. Using an unsupervised machine learning approach, we created BPD subclasses (e.g., endotypes) by clustering whole microarray data. T helper 17 cell differentiation was the most significant pathway differentiating the BPD endotypes.IntroductionBronchopulmonary dysplasia (BPD) is the most common complication of extreme prematurity. Discovery of BPD endotypes in an unbiased format, derived from the peripheral blood transcriptome, may uncover patterns underpinning this complex lung disease.MethodsAn unsupervised agglomerative hierarchical clustering approach applied to genome-wide expression of profiling from 62 children at day of life five was used to identify BPD endotypes. To identify which genes were differentially expressed across the BPD endotypes, we formulated a linear model based on least-squares minimization with empirical Bayes statistics.ResultsFour BPD endotypes (A, B,C,D) were identified using 7,319 differentially expressed genes. Across BPD endotypes, 5,850 genes had a p value < 0.05 after multiple comparison testing. Endotype A consisted of neonates with a higher gestational age and birthweight. Endotypes B-D included neonates between 25 and 26 weeks and a birthweight range of 640 to 940 g. Endotype D appeared to have a protective role against BPD compared to Endotypes B and C (36% vs. 62% vs. 60%, respectively). The most significant pathway focused on T helper 17 cell differentiation.ConclusionBioinformatic analyses can help identify BPD endotypes that associate with clinical definitions of BPD.