Raman Spectroscopy Combined with Machine Learning Reveals Myalgic Encephalomyelitis–Associated Biomolecular Signatures at Rest and After Standardized Stress
Abstract
Myalgic encephalomyelitis (ME) is characterized by profound fatigue, post-exertional malaise (PEM), and cognitive dysfunction. Despite its clinical significance, the pathophysiology of PEM and disease heterogeneity remain unclear, and no validated biomarkers are available for rapid diagnosis or monitoring.
We aimed to develop a screening approach combining label-free Raman spectroscopy (RS) and machine learning modeling (ML) to detect biomolecular changes in blood plasma and differentiate patients with ME from sedentary healthy controls.
Blood plasma was collected from 115 patients with ME and 45 controls at rest (T0) and 90 min after a standardized, non-invasive stress test designed to induce PEM.
Plasma samples were analyzed by RS, and ML models were developed independently at each time point to differentiate patients with ME and controls. The RS-ML models identified spectral features consistent with contributions from proteins, lipids, and low-molecular-weight metabolites.
At T0 and T90, the area under the receiver operating characteristic curve, accuracy, specificity and sensitivity were 0.85 and 0.83, 79% and 84%, 82% and 90%, and 73% and 69%, respectively.
RS-ML provides a rapid, low-cost approach to detect ME-associated biomolecular signatures in plasma and capture biochemical alterations associated with standardized stress.
Web | DOI | PMC | PDF | International Journal of Molecular Sciences | Open Access
Successfully differentiates patients from sedentary healthy controls. Is there reason to think this would work if controls included non-healthy controls from similar illnesses? Why not include them here?
Successfully differentiates patients from sedentary healthy controls. Is there reason to think this would work if controls included non-healthy controls from similar illnesses? Why not include them here?
Probably expensive, also we don’t really know of illnesses that are similar.
The best case control match you can get is sedentary controls from NASA testing where they have studied bed rest for sending people to space. I think Rob Wust had access to these controls, obviously this is extremely expensive and limited supply.
Maybe "similar illnesses" was not the right wording. I think I've seen diabetes or MS patients used as controls in similar studies. I'd argue including people with depression or chronic fatigue without PEM would make the same result a lot more interesting.
Without non-healthy, non-MECFS controls, can we conclude anything beyond this method successfully differentiates healthy from non-healthy?
In the acknowledgements I don't see any mention of the Oxford team or Dr. Xu who did the pioneering work an Raman in ME/CFS. I thought that a bit strange as it would have made perfect sense to start from best practices and talk to the original authors and build on their work. After all, OMF and OMF research center directors like to promote themselves as good collaborators.
The problem isn’t to identify people that look like they might have ME/CFS. That’s very easy.
The problem is to figure out if they have ME/CFS or something else, or both.
If they can get something useful out of this work it would be great, but it feels like yet another fishing expedition. The biggest benefit might be that the field could be moving away from CPETs to these kind of passive massage stimuli tests, or the thumb tests Fluge and Mella are working on.
I can understand the push for a diagnostic test, patients want it to prove something which is otherwise disbelieved so charities push for it. But I don’t think having a test alone would change or improve things for us much. And when you add in sensitivity snd specificity you just introduce more problems over who can legitimise say they have a condition.
What may be more interesting is what using a technique like this may be able to tell us about differences and therefore potential mechanisms. That could be worth exploring more in this and the other raman spectroscopy studies. If something reproducible can be done and then dug into more.
Also agree with @Utsikt thst the passive nature may be an interesting one. Although I guess it raises questions over what you’re measuring, can you passively produce PEM or if you’re avoiding PEM what are you measuring, or what does that tell us?
Hello all. I am concerned about the machine learning practice in this paper, but do not have the capacity to dig in.
From a cursory glance, there appears to be no validation split of the data? I could have misread while skimming (fog), but the (many) hyperparameters would therefore have been optimised on the test sets of the 5-fold splits. This is poor practice and means model performance will be an overestimate of real life performance on a new cohort.
Additionally, I am not a fan of the random splitting, but may have missed further comment on why it was done. This leads to the situation where the model may, for every test sample, have seen a train sample from the same input space cluster (rather than being able to generalise across input space clusters). Again, this would lead to overestimating real performance. Is there a table somewhere reporting performance on all folds of the final model version, or some other model version? This would clarify performance variance.
For context, I have a PhD in AI (partly using health records) and currently research AI for drug discovery for a big tech company (off sick with ME).
Also agree with @Utsikt thst the passive nature may be an interesting one. Although I guess it raises questions over what you’re measuring, can you passively produce PEM or if you’re avoiding PEM what are you measuring, or what does that tell us?
I forgot to respond to this. It think it could possibly tell us if there are «background» alterations into something related to physical exertion. If PEM is purely caused by e.g. neurons misinterpreting normal peripheral signals, a sub-PEM test wouldn’t show anything in the periphery. Unless PEM has a lasting effect on the periphery beyond the flare.
The issue might be controls as always. Lots of things probably effect the things we are trying to measure in the periphery.
Hello all. I am concerned about the machine learning practice in this paper, but do not have the capacity to dig in.
From a cursory glance, there appears to be no validation split of the data? I could have misread while skimming (fog), but the (many) hyperparameters would therefore have been optimised on the test sets of the 5-fold splits. This is poor practice and means model performance will be an overestimate of real life performance on a new cohort.
Additionally, I am not a fan of the random splitting, but may have missed further comment on why it was done. This leads to the situation where the model may, for every test sample, have seen a train sample from the same input space cluster (rather than being able to generalise across input space clusters). Again, this would lead to overestimating real performance. Is there a table somewhere reporting performance on all folds of the final model version, or some other model version? This would clarify performance variance.
For context, I have a PhD in AI (partly using health records) and currently research AI for drug discovery for a big tech company (off sick with ME).
Welcome and Thank you catching that. I do not understand how such issues pass from the review process.
You are right, this kind of cross validation is way too optimistic. I did not have too much time to see the paper but from a quick look I see an issue to the values used for specificity and sensitivity between those shown on table 3 and the confusion matrix (sensitivity and specificity appears to be swapped). Have they used HCs as the positive class?
Also potential issue with Feature selection leakage because it is not clear whether FS took place inside the cross validation loop.
EDIT : I also see a problem on how they are attempting to correct the class imbalance. They are not using equal weights per class and they effectively inverted the balancing.
I can understand the push for a diagnostic test, patients want it to prove something which is otherwise disbelieved so charities push for it. But I don’t think having a test alone would change or improve things for us much. And when you add in sensitivity snd specificity you just introduce more problems over who can legitimise say they have a condition.
I used to think a biomarker should be a research priority. I thought it would make diagnosis faster and validate our disease, but I no longer think so. It seems the diagnostic delay is mainly driven by a lack of doctors knowledgeable in our disease. And then, for the validation piece, this interview with Dr. Luis Nacul from 2019 gave me a new perspective. Here’s a quote that sums up his point quite nicely (my emphasis added):
I know quite a few doctors who know about ME and we can talk about ME in a unified language. But many do not understand it and so there is still a stigma about ME. As long as doctors will not take this illness seriously, the stigma will not be resolved. The current paradigm in medicine is that to prove something, you need to show a biomarker, and that’s why I think it’s so important to find one. But really, we don’t need a biomarker. We don’t have a biomarker for migraines, and yet we have legitimized it as an illness. So this should also be possible for ME.
I used to think a biomarker should be a research priority. I thought it would make diagnosis faster and validate our disease, but I no longer think so. It seems the diagnostic delay is mainly driven by a lack of doctors knowledgeable in our disease. And then, for the validation piece, this interview with Dr. Luis Nacul from 2019 gave me a new perspective. Here’s a quote that sums up his point quite nicely (my emphasis added):
Useable biomarkers are also based on pathology. The ones that aren’t will be calibrated against diagnoses based on clinical assessments, so they can’t surpass those in accuracy.
And any understanding of pathology will solve the perceived validity issue much better than a non-pathological test, so we might as well shoot for pathological understanding if we’re going to prioritise something.
I’m in a similar position to you I think @SugarSquared its something my views have changed on. I hadn’t seen that quote before (or had but forgot it!) but really like it. Thanks for sharing!
La maladie à coronavirus 2019 (COVID-19), causée par le coronavirus du syndrome respiratoire aigu sévère 2 (SARS-CoV-2), a entraîné plus de 7 millions de décès dans le monde, tandis que le COVID long affecte encore environ 400 millions de personnes. De nombreux symptômes du COVID long...
umontreal.scholaris.ca
Combining label-free Raman spectroscopy with machine learning for biomarker discovery in infectious and chronic diseases
La maladie à coronavirus 2019 (COVID-19), causée par le coronavirus du syndrome respiratoire aigu sévère 2 (SARS-CoV-2), a entraîné plus de 7 millions de décès dans le monde, tandis que le COVID long affecte encore environ 400 millions de personnes. De nombreux symptômes du COVID long chevauchent ceux de l’encéphalomyélite myalgique (EM), un trouble chronique et multisystémique touchant environ 600 000 Canadiens et jusqu’à 2,5 millions d’Américains. La COVID-19 et l’EM sont toutes deux des maladies hétérogènes, dont les manifestations cliniques vont de symptômes légers à modérés à des formes sévères, voire potentiellement mortelles. À l’heure actuelle, les tests diagnostiques de référence pour la COVID-19 présentent des limites quant à leur capacité à stratifier les patients selon les différents stades de la maladie ou à assurer un suivi à long terme. Par ailleurs, l’EM ne dispose d’aucun biomarqueur validé à des fins diagnostiques ou pronostiques, et les mécanismes moléculaires sous-jacents à cette maladie demeurent mal compris. Ces limitations entravent le développement d’outils diagnostiques et de suivi rapides au point de service, capables d’identifier les altérations biochimiques induites par l’infection au SARS-CoV-2 et par l’EM. Combler ces lacunes pourrait fournir des connaissances essentielles pour améliorer la prise en charge clinique et favoriser le développement de nouvelles stratégies thérapeutiques. L’objectif de cette thèse était de développer un outil de dépistage combinant la spectroscopie Raman sans marquage (RS) et la modélisation par apprentissage automatique (ML) (RS-ML) afin de permettre la détection sensible des altérations biochimiques associées aux maladies induites par l’infection au SARS-CoV-2 et l’EM, à différents stades des maladies aiguës et chroniques. Cette thèse comprend trois études visant à caractériser les signatures biomoléculaires dans le plasma sanguin humain associées à des conditions infectieuses et chroniques. Dans la première étude, un outil de dépistage RS-ML a été développé pour prédire la sévérité et la mortalité de la COVID-19 chez des patients hospitalisés. Les modèles de classification ont atteint des valeurs d’AUC comprises entre 0,83 et 0,94, indiquant que RS-ML pourrait aider à identifier les patients présentant un risque accru de complications ou de décès. Dans la deuxième étude, l’approche RS-ML a été appliquée pour détecter les changements biomoléculaires longitudinaux de la phase aiguë à la phase de récupération chez des patients hospitalisés atteints de COVID-19, et pour les différencier de sujets sains. Les modèles ont permis de distinguer avec succès les patients critiques des non-critiques pendant les phases aigües et de récupération, et d’autres modèles ont classé l’état des patients (aigu vs récupération) à l’aide d’analyses transversales et longitudinales. Ces modèles ont atteint des valeurs d’AUC comprises entre 0,83 et 1,00, démontrant le potentiel de RS-ML pour suivre l’évolution de la maladie et la récupération. Dans la troisième étude, la méthodologie RS-ML a été appliquée pour différencier les patients atteints d’EM des sujets sains à deux moments : au départ avant un test de stress (T0) et 90 minutes après un test de stress post-effort (T90). Les modèles ont atteint des valeurs d’AUC de 0,83 à T0 et 0,84 à T90, mettant en évidence la capacité de RS-ML à capter les altérations biochimiques induites par les stress associés au malaise post-effort. En conclusion, les résultats présentés dans cette thèse démontrent que l’intégration de la spectroscopie Raman sans marquage et de l’apprentissage automatique permet d’identifier des différences biomoléculaires spécifiques à la maladie, tant pour la COVID-19 que pour l’EM, à différents stades de la maladie. Ces résultats soutiennent le potentiel de l’approche RS-ML comme outil de stratification des patients et de suivi clinique.
Coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has resulted in over 7 million deaths worldwide, while long COVID continues to affect an estimated 400 million individuals. Many symptoms of long COVID overlap with those of myalgic encephalomyelitis (ME), a chronic, multisystem disorder affecting approximately 600,000 Canadians and up to 2.5 million Americans. Both COVID-19 and ME are heterogeneous conditions, with clinical manifestations ranging from mild to moderate symptoms to severe and potentially life-threatening disease.
Currently, gold-standard diagnostic tests for COVID-19 are limited in their ability to stratify patients across different stages of disease or to support long-term monitoring. Moreover, ME lacks validated biomarkers for diagnosis or prognosis, and the molecular mechanisms underlying the disease remain poorly understood. These limitations hinder the development of rapid point-of-care diagnostic and monitoring tools capable of identifying biochemical alterations induced by SARS-CoV-2 infection and ME. Addressing these gaps could provide critical insights to improve clinical management and support the development of novel therapeutic strategies.
The aim of this thesis was to develop a screening tool that combines label-free Raman spectroscopy (RS) with machine learning (ML) modeling (RS-ML) to enable sensitive detection of disease-associated biochemical alterations induced by SARS-CoV-2 infection and ME across different stages of acute and chronic disease. This thesis comprises three studies designed to characterize biomolecular signatures in human blood plasma associated with infectious and chronic conditions.
In the first study, an RS-ML screening tool was developed to predict COVID-19 severity and mortality in hospitalized patients. The classification models achieved AUC values ranging from 0.83 to 0.94, indicating that RS-ML may help identify patients at increased risk of complications or death.
In the second study, the RS-ML approach was applied to detect longitudinal biomolecular changes from acute infection to recovery in hospitalized COVID-19 patients and to differentiate these patients from healthy controls. The models successfully distinguished critical from non-critical patients during both acute and recovery phases, and additional models classified patient status (acute vs. recovery) using cross-sectional and 6 longitudinal analyses. These models achieved AUC values ranging from 0.83 to 1.00, demonstrating the potential of RS-ML to monitor disease progression and recovery.
In the third study, the RS-ML methodology was applied to differentiate ME patients from healthy controls at two time points: baseline prior to a stress test (T0) and 90 minutes following the post-exertional stress test (T90). The models achieved AUC values of 0.83 at T0 and 0.84 at T90, highlighting the ability of RS-ML to capture stress-induced biomolecular alterations associated with post-exertional malaise.
In conclusion, the findings presented in this dissertation demonstrate that the integration of label-free Raman spectroscopy with machine learning enables the identification of disease-specific biomolecular differences in both COVID-19 and ME across distinct stages of disease. These results support the potential of RS-ML as a tool for disease stratification and clinical monitoring.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
