Tag: machine learning

Discriminating Myalgic Encephalomyelitis/Chronic Fatigue Syndrome and comorbid conditions using metabolomics in UK Biobank

Abstract:

Background: Diagnosing complex illnesses like Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) is complicated due to the diverse symptomology and presence of comorbid conditions. ME/CFS patients often present with multiple health issues, therefore, incorporating comorbidities into research can provide a more accurate understanding of the condition’s symptomatology and severity, to better reflect real-life patient experiences.

Methods: We performed association studies and machine learning on 1194 ME/CFS individuals with blood plasma nuclear magnetic resonance (NMR) metabolomics profiles, and seven exclusive comorbid cohorts: hypertension (n = 13,559), depression (n = 2522), asthma (n = 6406), irritable bowel syndrome (n = 859), hay fever (n = 3025), hypothyroidism (n = 1226), migraine (n = 1551) and a non-diseased control group (n = 53,009).

Results: We present a lipoprotein perspective on ME/CFS pathophysiology, highlighting gender-specific differences and identifying overlapping associations with comorbid conditions, specifically surface lipids, and ketone bodies from 168 significant individual biomarker associations. Additionally, we searched for, trained, and optimised a machine learning algorithm, resulting in a predictive model using 19 baseline characteristics and nine NMR biomarkers which could identify ME/CFS with an AUC of 0.83 and recall of 0.70. A multi-variable score was subsequently derived from the same 28 features, which exhibited ~2.5 times greater association than the top individual biomarker.

Conclusions: This study provides an end-to-end analytical workflow that explores the potential clinical utility that association scores may have for ME/CFS and other difficult to diagnose conditions.

Source: Huang K, G C de Sá A, Thomas N, Phair RD, Gooley PR, Ascher DB, Armstrong CW. Discriminating Myalgic Encephalomyelitis/Chronic Fatigue Syndrome and comorbid conditions using metabolomics in UK Biobank. Commun Med (Lond). 2024 Nov 26;4(1):248. doi: 10.1038/s43856-024-00669-7. PMID: 39592839; PMCID: PMC11599898.  https://pmc.ncbi.nlm.nih.gov/articles/PMC11599898/ (Full text)

Fast Targeted Metabolomics for Analyzing Metabolic Diversity of Bacterial Indole Derivatives in ME/CFS Gut Microbiome

Abstract:

Disruptions in microbial metabolite interactions due to gut microbiome dysbiosis and metabolomic shifts may contribute to Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) and other immune-related conditions. The aryl hydrocarbon receptor (AhR), activated upon binding various tryptophan metabolites, modulates host immune responses. This study investigates whether the metabolic diversity-the concentration distribution-of bacterial indole pathway metabolites can differentiate bacterial strains and classify ME/CFS samples.

A fast targeted liquid chromatography-parallel reaction monitoring method at a rate of 4 minutes per sample was developed for large-scale analysis. This method revealed significant metabolic differences in indole derivatives among B. uniformis strains cultured from human isolates. Principal component analysis identified two major components (PC1, 68.9%; PC2, 18.7%), accounting for 87.6% of the variance and distinguishing two distinct B. uniformis clusters. The metabolic difference between clusters was particularly evident in the relative contributions of indole-3-acrylate and indole-3-aldehyde.

We further measured concentration distributions of indole derivatives in ME/CFS by analyzing fecal samples from 10 patients and 10 healthy controls using the fast targeted metabolomics method. An AdaBoost-LOOCV model achieved moderate classification success with a mean LOOCV accuracy of 0.65 (Control: precision of 0.67, recall of 0.60, F1-score of 0.63; ME/CFS: precision of 0.64, recall of 0.7000, F1-score of 0.67).

These results suggest that the metabolic diversity of indole derivatives from tryptophan degradation, facilitated by the fast targeted metabolomics and machine learning, is a potential biomarker for differentiating bacterial strains and classifying ME/CFS samples.

Mass spectrometry datasets are accessible at the National Metabolomics Data Repository (ST002308, DOI: 10.21228/M8G13Q; ST003344, DOI: 10.21228/M8RJ9N; ST003346, DOI: 10.21228/M8RJ9N).

Source: Tian H, Wang L, Aiken E, Ortega RJV, Hardy R, Placek L, Kozhaya L, Unutmaz D, Oh J, Yao X. Fast Targeted Metabolomics for Analyzing Metabolic Diversity of Bacterial Indole Derivatives in ME/CFS Gut Microbiome. bioRxiv [Preprint]. 2024 Jul 29:2024.07.29.605643. doi: 10.1101/2024.07.29.605643. PMID: 39131327; PMCID: PMC11312560. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11312560/ (Full text)

Development of an expert system for the classification of myalgic encephalomyelitis/chronic fatigue syndrome

Abstract:

Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a severe condition with an uncertain origin and a dismal prognosis. There is presently no precise diagnostic test for ME/CFS, and the diagnosis is determined primarily by the presence of certain symptoms. The current study presents an explainable artificial intelligence (XAI) integrated machine learning (ML) framework that identifies and classifies potential metabolic biomarkers of ME/CFS.

Metabolomic data from blood samples from 19 controls and 32 ME/CFS patients, all female, who were between age and body mass index (BMI) frequency-matched groups, were used to develop the XAI-based model. The dataset contained 832 metabolites, and after feature selection, the model was developed using only 50 metabolites, meaning less medical knowledge is required, thus reducing diagnostic costs and improving prognostic time. The computational method was developed using six different ML algorithms before and after feature selection. The final classification model was explained using the XAI approach, SHAP.

The best-performing classification model (XGBoost) achieved an area under the receiver operating characteristic curve (AUCROC) value of 98.85%. SHAP results showed that decreased levels of alpha-CEHC sulfate, hypoxanthine, and phenylacetylglutamine, as well as increased levels of N-delta-acetylornithine and oleoyl-linoloyl-glycerol (18:1/18:2)[2], increased the risk of ME/CFS. Besides the robustness of the methodology used, the results showed that the combination of ML and XAI could explain the biomarker prediction of ME/CFS and provided a first step toward establishing prognostic models for ME/CFS.

Source: Yagin FH, Shateri A, Nasiri H, Yagin B, Colak C, Alghannam AF. Development of an expert system for the classification of myalgic encephalomyelitis/chronic fatigue syndrome. PeerJ Comput Sci. 2024 Mar 20;10:e1857. doi: 10.7717/peerj-cs.1857. PMID: 38660205; PMCID: PMC11041999. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11041999/ (Full text)

Exploring Cognitive Dysfunction in Long COVID Patients: Eye Movement Abnormalities and Frontal-Subcortical Circuits Implications via Eye-Tracking and Machine Learning

Abstract:

Background: Cognitive dysfunction is regarded as one of the most severe aftereffects following coronavirus disease 2019 (COVID-19). Eye movements, controlled by various brain regions, including the dorsolateral prefrontal cortex and frontal-thalamic circuits, offer a potential metric for evaluating cognitive dysfunction. We aimed to examine the utility of eye movement measurements in identifying cognitive impairments in long COVID patients.

Methods: We recruited 40 long COVID patients experiencing subjective cognitive complaints and 40 healthy controls and used a certified eye-tracking medical device to record saccades and antisaccades. Machine learning was applied to enhance the analysis of eye movement data.

Results: Patients did not differ from the healthy controls regarding age, sex, and years of education. However, the patients’ Montreal Cognitive Assessment total score was significantly lower than healthy controls. Most eye movement parameters were significantly worse in patients: the latencies, gain, and velocity of visually and memory-guided saccades, the number of correct memory saccades, the latencies and duration of reflexive saccades, and the number of errors in the antisaccade test. Machine learning permitted distinguishing between long COVID patients experiencing subjective cognitive complaints and healthy controls.

Conclusion: Our findings suggest impairments in frontal subcortical circuits in long COVID patients experiencing subjective cognitive complaints. Eye-tracking, combined with machine learning, offers a novel, efficient way to assess and monitor long COVID patients’ cognitive dysfunctions, suggesting its utility in clinical settings for early detection and personalized treatment strategies. Further research is needed to determine the long-term implications of these findings and the reversibility of cognitive dysfunctions.

Source: Benito-León J, Lapeña J, García-Vasco L, Cuevas C, Viloria-Porto J, Calvo-Córdoba A, Arrieta-Ortubay E, Ruiz-Ruigómez M, Sánchez-Sánchez C, García-Cena C. Exploring Cognitive Dysfunction in Long COVID Patients: Eye Movement Abnormalities and Frontal-Subcortical Circuits Implications via Eye-Tracking and Machine Learning. Am J Med. 2024 Apr 5:S0002-9343(24)00217-1. doi: 10.1016/j.amjmed.2024.04.004. Epub ahead of print. PMID: 38583751. https://pubmed.ncbi.nlm.nih.gov/38583751/

Machine learning algorithms for detection of visuomotor neural control differences in individuals with PASC and ME

Abstract:

The COVID-19 pandemic has affected millions worldwide, giving rise to long-term symptoms known as post-acute sequelae of SARS-CoV-2 (PASC) infection, colloquially referred to as long COVID. With an increasing number of people experiencing these symptoms, early intervention is crucial. In this study, we introduce a novel method to detect the likelihood of PASC or Myalgic Encephalomyelitis (ME) using a wearable four-channel headband that collects Electroencephalogram (EEG) data. The raw EEG signals are processed using Continuous Wavelet Transform (CWT) to form a spectrogram-like matrix, which serves as input for various machine learning and deep learning models. We employ models such as CONVLSTM (Convolutional Long Short-Term Memory), CNN-LSTM, and Bi-LSTM (Bidirectional Long short-term memory). Additionally, we test the dataset on traditional machine learning models for comparative analysis.

Our results show that the best-performing model, CNN-LSTM, achieved an accuracy of 83%. In addition to the original spectrogram data, we generated synthetic spectrograms using Wasserstein Generative Adversarial Networks (WGANs) to augment our dataset. These synthetic spectrograms contributed to the training phase, addressing challenges such as limited data volume and patient privacy. Impressively, the model trained on synthetic data achieved an average accuracy of 93%, significantly outperforming the original model.

These results demonstrate the feasibility and effectiveness of our proposed method in detecting the effects of PASC and ME, paving the way for early identification and management of the condition. The proposed approach holds significant potential for various practical applications, particularly in the clinical domain. It can be utilized for evaluating the current condition of individuals with PASC or ME, and monitoring the recovery process of those with PASC, or the efficacy of any interventions in the PASC and ME populations. By implementing this technique, healthcare professionals can facilitate more effective management of chronic PASC or ME effects, ensuring timely intervention and improving the quality of life for those experiencing these conditions.

Source: Harit Ahuja, Smriti Badhwar, Heather Edgell, Lauren E. Sergio, Marin Litoiu. Machine learning algorithms for detection of visuomotor neural control differences in individuals with PASC and ME. Front. Hum. Neurosci. Sec. Brain-Computer Interfaces, Volume 18 – 2024 | doi: 10.3389/fnhum.2024.1359162 https://www.frontiersin.org/articles/10.3389/fnhum.2024.1359162/full (Full text)

Sex differences in symptomatology and immune profiles of Long COVID

Abstract:

Strong sex differences in the frequencies and manifestations of Long COVID (LC) have been reported with females significantly more likely than males to present with LC after acute SARS-CoV-2 infection1-7. However, whether immunological traits underlying LC differ between sexes, and whether such differences explain the differential manifestations of LC symptomology is currently unknown.

Here, we performed sex-based multi-dimensional immune-endocrine profiling of 165 individuals8 with and without LC in an exploratory, cross-sectional study to identify key immunological traits underlying biological sex differences in LC.

We found that female and male participants with LC experienced different sets of symptoms, and distinct patterns of organ system involvement, with female participants suffering from a higher symptom burden. Machine learning approaches identified differential sets of immune features that characterized LC in females and males. Males with LC had decreased frequencies of monocyte and DC populations, elevated NK cells, and plasma cytokines including IL-8 and TGF-β-family members.

Females with LC had increased frequencies of exhausted T cells, cytokine-secreting T cells, higher antibody reactivity to latent herpes viruses including EBV, HSV-2, and CMV, and lower testosterone levels than their control female counterparts. Testosterone levels were significantly associated with lower symptom burden in LC participants over sex designation.

These findings suggest distinct immunological processes of LC in females and males and illuminate the crucial role of immune-endocrine dysregulation in sex-specific pathology.

Source: Julio Silva, Takehiro Takahashi, Jamie Wood, Peiwen Lu, Sasha Tabachnikova, Jeffrey Gehlhausen, Kerrie Greene, Bornali Bhattacharjee, Valter Silva Monteiro, Carolina Lucas, Rahul Dhodapkar, Laura Tabacof, Mario Pena-Hernandez, Kathy Kamath, Tianyang Mao, Dayna Mccarthy, Ruslan Medzhitov, David van Dijk, Harlan Krumholz, Leying Guan, David Putrino, Akiko Iwasaki. Sex differences in symptomatology and immune profiles of Long COVID. medRxiv 2024.02.29.24303568; doi: https://doi.org/10.1101/2024.02.29.24303568 https://www.medrxiv.org/content/10.1101/2024.02.29.24303568v1 (Full study available as PDF file)

Long COVID Diagnostic with Differentiation from Chronic Lyme Disease using Machine Learning and Cytokine Hubs

Abstract:

The absence of a diagnostic for long COVID (LC) or post-acute sequelae of COVID-19 (PASC) has profound implications for research and potential therapeutics. Further, symptom-based identification of patients with long-term COVID-19 lacks the specificity to serve as a diagnostic because of the overlap of symptoms with other chronic inflammatory conditions like chronic Lymedisease (CLD), myalgic encephalomyelitis-chronic fatigue syndrome (ME-CFS), and others. Here, we report a machine-learning approach to long COVID diagnosis using cytokine hubs that are also capable of differentiating long COVID from chronic Lyme.

We constructed three tree-based classifiers: decision tree, random forest, and gradient-boosting machine (GBM) and compared their diagnostic capabilities. A 223 patient dataset was partitioned into training (178 patients) and evaluation (45 patients) sets. The GBM model was selected based on performance (89% Sensitivity and 96% Specificity for LC) with no evidence of overfitting.

We tested the GBM on a random dataset of 124 individuals (106 PASC and 18 Lyme), resulting in high sensitivity (97%) and specificity 90% for LC). A Lyme Index composed of two features ((TNF-alpha +IL-4)/(IFN-gamma + IL-2) and (TNF-alpha *IL-4)/(IFN-gamma + IL-2 + CCL3) was constructed as a confirmatory algorithm to discriminate between LC and CLD.

Source: Bruce Patterson, Jose Guevara-Coto, Javier Mora et al. Long COVID Diagnostic with Differentiation from Chronic Lyme Disease using Machine Learning and Cytokine Hubs, 18 January 2024, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-3873244/v1] https://www.researchsquare.com/article/rs-3873244/v1 (Full text)

Psychometric evaluation of the DePaul Symptom Questionnaire-Short Form (DSQ-SF) among adults with Long COVID, ME/CFS, and healthy controls: A machine learning approach

Abstract:

Long COVID shares a number of clinical features with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), including post-exertional malaise, severe fatigue, and neurocognitive deficits. Utilizing validated assessment tools that accurately and efficiently screen for these conditions can facilitate diagnostic and treatment efforts, thereby improving patient outcomes.

In this study, we generated a series of random forest machine learning algorithms to evaluate the psychometric properties of the DePaul Symptom Questionnaire-Short Form (DSQ-SF) in classifying large groups of adults with Long COVID, ME/CFS (without Long COVID), and healthy controls.

We demonstrated that the DSQ-SF can accurately classify these populations with high degrees of sensitivity and specificity. In turn, we identified the particular DSQ-SF symptom items that best distinguish Long COVID from ME/CFS, as well as those that differentiate these illness groups from healthy controls.

Source: McGarrigle WJ, Furst J, Jason LA. Psychometric evaluation of the DePaul Symptom Questionnaire-Short Form (DSQ-SF) among adults with Long COVID, ME/CFS, and healthy controls: A machine learning approach. J Health Psychol. 2024 Jan 28:13591053231223882. doi: 10.1177/13591053231223882. Epub ahead of print. PMID: 38282368. https://pubmed.ncbi.nlm.nih.gov/38282368/

Persistent complement dysregulation with signs of thromboinflammation in active Long Covid

Abstract:

Long Covid is a debilitating condition of unknown etiology. We performed multimodal proteomics analyses of blood serum from COVID-19 patients followed up to 12 months after confirmed severe acute respiratory syndrome coronavirus 2 infection. Analysis of >6500 proteins in 268 longitudinal samples revealed dysregulated activation of the complement system, an innate immune protection and homeostasis mechanism, in individuals experiencing Long Covid.

Thus, active Long Covid was characterized by terminal complement system dysregulation and ongoing activation of the alternative and classical complement pathways, the latter associated with increased antibody titers against several herpesviruses possibly stimulating this pathway. Moreover, markers of hemolysis, tissue injury, platelet activation, and monocyte–platelet aggregates were increased in Long Covid. Machine learning confirmed complement and thromboinflammatory proteins as top biomarkers, warranting diagnostic and therapeutic interrogation of these systems.

Source: Carlo Cervia-Hasler et al. Persistent complement dysregulation with signs of thromboinflammation in active Long Covid. Science383,eadg7942(2024). DOI: 10.1126/science.adg7942 https://www.science.org/doi/10.1126/science.adg7942 (Full text)

Data-driven prognosis of long COVID in patients using machine learning

Abstract:

Long-COVID is a health condition in which individuals experience persisting, returning or new symptoms longer than 4 weeks after they have recovered from COVID-19 and this condition can even last for months. It can cause multi-organ failure and in some cases, it can even lead to death. The effects and symptoms of Long COVID can vary from person to person. Even though it’s rising globally, there is a limited understanding about its prediction, risk factors and whether its prognosis can be predicted in the initial first week of acute COVID-19. Artificial Intelligence (AI) and Machine Learning (ML) have aided the medical industry in a variety of ways including the diagnosis, prediction, and prognosis of many diseases.

This paper introduces a novel method to determine Long COVID in the early or first week of acute COVID-19 by considering the basic demographics, and symptoms during COVID-19, along with the clinical lab results of the patients hospitalized. In comparison with different ML models such as Logistic Regression, Support Vector Machine (SVM), XGBoost and Artificial Neural Network (ANN) to predict and classify the patients as Long COVID or Short COVID during the first week of COVID-19, ANN has outperformed the other models with an accuracy of 81% when considering the symptoms of COVID-19 and a 79% for the clinical test data. The predictive factors and the significant clinical tests for the Long COVID are also determined by using different methods like Chi-square Test and Pearson Correlation.

Source: S. S. Parvathy, Nagesh Subbanna, Sethuraman Rao, Rahul Krishnan Pathinarupothi, T. S. Dipu, Merlin Moni, Chithira V. Nair; Data-driven prognosis of long COVID in patients using machine learning. AIP Conf. Proc. 15 December 2023; 2901 (1): 060014. https://doi.org/10.1063/5.0178561 https://pubs.aip.org/aip/acp/article/2901/1/060014/2930006 (Full text available as PDF file)