Development of PI-ML Algorithm for Prediction of the Real-time Risk for Developing Pre-diabetes

Body Weight Clinical Trial

— PRAESIIDIUM  

Official title:

Physics Informed Machine Learning-based Prediction and Reversion of Impaired Fasting Glucose Management

Clinical Trial Details — Status: Not yet recruiting

Administrative data

• NCT number: NCT06195566
• Other study ID #: 101095672
• Status: Not yet recruiting
• Start date: January 29, 2024
• Est. completion date: March 31, 2025

Study information

• Verified date: December 2023
• Source: University of Latvia
• Contact: Zane Smite, Mg. Sci
• Phone: (+371) 28761602
• Email: zane.smite[@]lu.lv
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

In this prospective, non-randomized, monocentric study, data will be collected from otherwise healthy individuals with overweight/obese grade I to increase data availability in the pre-diabetes field (impaired glucose intolerance), and to validate the outputs of an algorithm for the "physics-informed machine learning (PIML)" designed to estimate the real-time risk of prediabetes. Each participant will take part in the study for 4 months, including 3 onsite visits.

During the screening visit, participants' eligibility will be determined by checking the inclusion and exclusion criteria after detailed information and obtaining informed consent by the investigator. Blood will be withdrawn for exclusion of existing prediabetes/diabetes at the fasted state. For women in reproductive age, a urinary pregnancy test will be performed. After getting the results of blood tests (glucose and HbA1c), participants will be asked to participate in study.

On the visit 1, eligible participants will arrive at the study centre in a fasting state. Blood samples will be collected and participants will get vials and instructions for collection of stool and urine samples. Anthropometric data, lifestyle habit (cigarette, alcohol consumption) and family history will be collected. A 6-minute walking test to determine VO2 max will then be performed. Participants will receive a blinded Abbott Libre Pro glucose sensor, which they will wear for the next 14-days. Further, participants will be provided with a Fitbit Charge 5 health and fitness wristband. For validation purposes some part of study participants will be kindly asked to test newly develop wrist-worn device (EDIBit). With the help of 24-hour food recall, study subjects will be trained by medical staff on how to correctly enter their food intake in the Study app for completion of digital 3-day food diaries. They will be asked to fill in the diaries for 3 days after study visit1 and 3 days before study visit2. They will also receive a food frequency questionnaire during visit1.

The second study visit will run nearly identical to study visit1 (except for food frequency questionnaire which will be omitted). During this visit, participants will receive information sheets on physical activity and dietary recommendations.

The third and last visit will run nearly identically to the study visit2, except that no new glucose sensor will be inserted and also stool samples will not be collected.

Description:

Noncommunicable diseases (NCDs) such as cancer, cardiovascular diseases, and diabetes represent 74% of the disease burden globally and are the major causes of preventable premature deaths. (1,2).

Diabetes is a chronic NCD characterized by elevated glucose blood levels. In 2021, the prevalence of diabetes in Europe was 1 in 11 adults (61 million), a figure projected to rise to 69 million by 2045. (3) The global management and treatment of diabetes cost 988 billion dollars in 2021. Despite these expenditures, diabetes remains the third leading cause of death, accounting for 6.7 million deaths, with 1.1 millions of these in Europe alone. Among the different types, type-2 diabetes (T2D) comprises 90% of total diabetes cases, primarily presenting in adulthood. (3,4) Risk factors for T2D include genetic predisposition, family history, metabolic syndrome, obesity, physical inactivity, age, and ethnicity. There are 541 million adults worldwide with Impaired Glucose Tolerance (IGT), a significant risk factor for T2D. (5,6) IGT and Impaired Fasting Glucose (IFG) represent intermediate conditions within the "healthy-to-T2D transition" and are symptoms of prediabetes. (7,8) Notably, prediabetes represents an early-stage condition that can be reversed. Studies show that T2D progression can be reduced by approximately 58% within three years through lifestyle modifications. Physical activity of 30-54 minutes at least 2-5 days per week is recommended, as well as a healthy diet. (9) Efforts have been made to develop non-invasive diabetes risk prediction models based on clinically available parameters. (10) The onset of T2D involves complex, multiscale mechanisms starting from molecular, tissue, and organ levels, leading to dysfunction in physiological processes. Chronic inflammatory biomarkers play a significant role in T2D pathogenesis. Recognizing this multi-level approach is a step towards personalized disease diagnosis. However, there remain challenges related to modelling the "healthy-to-prediabetes transition" from both case study and methodological perspectives. (5,13,14) The objective of this project is the development of a prototype tool, aimed at real-time prediction of prediabetic risk. This tool will incorporate a series of patient-specific mathematical models simulating metabolism, pancreas hormone production, microbiome metabolites, inflammatory processes, and immune system response. These models were initially developed during the FP7 MISSION-T2D project and further developed into the implementation of an integrated, multilevel, and patient-specific model, incorporating genetic, metabolic, and nutritional data for the simulation and prediction of metabolic and inflammatory processes in the onset and progression of T2D. (14-18) The prediction algorithm will utilize a "physics-informed machine learning" (PIML) approach, combining a comprehensive dataset from both existing and new clinical trials, with continuous data input through wearable sensors. (19) The final algorithm will be hosted on a web-based platform where both medical professionals and patients can input data from multiple sources.

A dedicated prospective observational study described in this application will be conducted in Latvia recruiting adult participants with metabolic risk factors - overweight and obesity grade I, for data collection purposes to validate the developed machine learning PIML algorithm for pre-diabetes real-time risk prediction.

Data collections has three main purposes: I. Input Data for the in-silico MT2D model:

The input of the simulations includes the following discrete starting parameters: gender (M/F); weight; height; number of sessions of physical activity (0, 1, 2, 3); duration of the bout of physical activity (30, 60, 90 min); intensity in terms of % VO2max (40, 60); 3 meals per day; in each meal are specified the carbohydrates (low, medium, high), proteins (low, medium, high) and fats (low, medium, high).

II. Validation of the MT2D outputs: Output numerical values of the simulations from the model include:

1) Inflammation markers (recorded every 8h): B-cells (B-1, B-2), PBL, TH (Th1, Th2, Th17, Treg), CTL, Treg, NK, MA, DC, EP, ADIP (number, volume), IgM, IgG (IgG1, IgG2), IC, IL-2, IL-12, IFN-g, IL-4, TNF-a, TGF-b, IL-10, IL-6, IL-18, IL-23, IFN-b, IL-1b, LPS, leptin;

a) Metabolic outcomes (recorded every minute): arterial concentrations (glu, pyr, lac, ala, gly, FFA, tgl, O2, CO2); organs (22 metabolites); hormones (insulin, glucagon, epinephrine);fasting glucose; rate of appearance (glucose, alanine, triglycerides);total daily energy balance; %VO2max; anthropometric measures: BW, BMI, fat mass, fat free mass.

III. Data for training/validation of the physics-informed machine learning (PIML) algorithm:

demographic Data; health-related Data; lifestyle Data (e.g., food consumption data and physical activity data); continuous ingestion through wearable sensors (Continuous Glucose Monitoring (CGM and tracker of physical activity e.g., Fitbit Charge 5, EDIBit.) These devices, such as smartwatches and fitness/activity trackers (e.g., Fitbit, Fibion, Apple Watch), are equipped with sensors that can track a variety of health metrics, including physical activity, heart rate, sleep patterns and increasingly also for glucose monitoring (non-invasive continuous glucose monitoring are still under development). (20,21) This data can be used to identify patterns and trends in a person's health, which can help with the early detection of diabetes/prediabetes. (22) Machine learning (ML) models show potential in enhancing early detection by analysing various risk factors and predicting outcomes. However, before these models are integrated into healthcare systems and clinical practices, they must be rigorously evaluated. One of the most robust methods for such evaluation is through external validation using longitudinal cohorts. (23-26) During the clinical study, participants may be supported by a digital assistant which can be used to make automatic voice calls to participants to collect data and provide user support and follow-up. The digital assistant uses pre-defined dialogues (designed by trial staff) and vocal responses from participants are recorded as text/data in the trial database. The function of the digital assistant is regarded as a complement to other methods for data capture such as questionnaires, wearables, messaging, regular phone calls.

A subset of the participant will receive a study-dedicated e-SIM (to avoid using another mobile phone), on which number the participant could be called by the digital assistant. The participants will be informed of the subscriber number used by the digital assistant, so that they can recognise its incoming calls and freely decide whether to accept or reject them, on a personal choice. Participants will be able to stop this automated service at any time.

During the clinical study, participants may be supported by a digital assistant to help people better comply with the clinical study.

A subset of the participant will receive a study-dedicated e-SIM (to avoid using another mobile phone), on which number the participant could be called by the digital assistant. The participants will be informed of the subscriber number used by the digital assistant, so that they can recognise its incoming calls and freely decide whether to accept or reject them, on a personal choice. Participants will be able to stop this automated service at any time.

Timeline and probands: The study will run for 15 months. During this period, 75 individuals will be followed for 4 months. The recruitment will take place from January 2024 until March 2025, parallel to the study period.

Subject identification: For each participant who has signed the Participant Informed Consent Form, the investigator must allocate a unique two letter and three-digit Participant Identification Number.

All documents, forms and data (including bio-materials - urine, blood and stool samples) files will be tagged with this Participant ID. Each participant, eligible and not eligible, will be documented in the Screening and Enrolment Log.

Data management: The data of the participants will be entered in an eCRFs. An eCRF will be provided by the project consortium partners Spindox Labs and CheckHealth. The eCRF will be maintained by staff at the University of Latvia. All study data will be captured in the eCRF and monitored by the monitor. All processes will be handled in accordance with standard operating procedures (SOPs).

Collected data from wearable sensors and remote-assisted questionnaires will be managed by the LinkWatch cloud-based storage for the entire duration of the project. This includes regular back-up procedures and security provisions in accordance with the GDPR and the CHK security policy. The security policy is based on the framework recommended by the Swedish Contingency Agency (Reg. 2016/679/UE), following ISO/IEEE 27000.

No paper forms of the questionnaires will be used, except informed form. Consent form will be stored in archive at University of Latvia.

Macronutrient food intake will be obtained through a three-day food diary. That will be provided in different ways - initializing the macronutrient database with finish Finelli database, using Open food facts for packaged food. Meanwhile participants will be asked to double information, writing manually food diary for three days.

In visit 1 and visit 3 food frequency questionnaire and in each visit, 24h recall will be obtained. Physical activity will be tracked using a commercially available FitBit Charge 5 tracker.

Between visit 1 and visit 2, specific recommendations will not be provided. This phase will serve as a data collection period to accurately assess participants' current lifestyle factors. To provide a higher diversity of collected data, recommendations for physical activity and a healthy diet will be provided to the participants at visit 2. These prescriptions will align with the relevant WHO guidelines and summarized in hand-out materials.

In patients with T2DM, regular exercise increases insulin sensitivity and secretion, improving glucose tolerance. It is noted that a single bout of exercise can enhance insulin sensitivity, while long-term exercise is required to improve pancreatic function in T2DM patients. Myokines are secreted factors from skeletal muscle, adipose tissue, liver, gut, etc proposed as cross-talk organ mediators involved in metabolic adaptations to exercise. IL-6, IL-2, IL-10, leptin, will be assessed for PIML validation purposes.

Blood samples from enrolled patients at the three cohort visits (Visit 1, Visit 2 and Visit 3) will be collected and initially processed to obtain plasma samples and temporarily stored (at -80°C) by the recruiting site (see below blood samples processing before shipment).

Plasma samples will be shipped to the project partner Italian Liver Foundation (FIF) in Basovizza, Trieste (Italy). At FIF, plasma samples will be stored at the at a temperature (-80°C) and access-controlled freezer for 10 years.

At FIF, batched samples will be unfrozen and further processed to determine plasma abundances of interleukins (IL-2, IL-6, IL-10, and leptin).

Blood will be collected for determination of clinical blood tests on the study site and for sample transfer to FIF for further biomarker analysis and preservation at the University of Latvia.

Participants will collect their faecal samples at home (together two samples - in first and second visit, respectively). The primary goal of faecal samples collection is to create a long-lasting PRAESIIDIUM biobank. In a subsequent moment, faecal samples might be processed for microbiome analysis. After arrival at local hospital sample should be frozen at -80°C until processing. Further processing - upon availability of additional funds (Fondazione Italiana Fegato ONLUS, Italy).

Recruitment information / eligibility

• Status: Not yet recruiting
• Enrollment: 75
• Est. completion date: March 31, 2025
• Est. primary completion date: March 31, 2025
• Accepts healthy volunteers: Accepts Healthy Volunteers
• Gender: All
• Age group: 18 Years to 65 Years

Eligibility

Inclusion Criteria:

• Healthy adult volunteers (age = 18 years old);

• Overweight (BMI 25 - 29.9 kg/m2) and obese grade I individuals (with BMI 30 - 34.9 kg/m2);

• Written consent of the participant after being informed;

• Ownership of a smartphone running Android or iOS.

Exclusion Criteria:

• Non-compliance;

• Ongoing treatment with immunosuppressive and/or anti-inflammatory medications (NSAIDs, glucocorticoids, chemotherapy, biologicals);

• Ongoing treatment with glucose lowering drugs, except if anti-diabetic medication has not been stopped - for metformin one month, for GLP-1 RA, tirzepatide - two months prior enrolment;

• Presence of autoimmune and/or inflammatory disease (autoimmune thyroid disease, psoriasis, inflammatory bowel disease);

• Skin conditions hindering application of continuous glucose monitoring systems;

• Diabetes or prediabetes as diagnosed by ADA/WHO criteria according to fasting glucose and/or HbA1c;

• High risk alcohol consumption - according to NIAAA - National Institute on Alcohol Abuse and Alcoholism (for men - more than 4 drinks on any day or more than 14 drinks per week; for women - more than 3 drinks on any day or more than 7 drinks per week);

• Factors otherwise limiting the participation in the study according to the judgement of the investigator;

• Pregnancy or intention to get pregnant during the study timeline.

Related Conditions & MeSH terms

• Body Weight
• Life Style, Healthy
• Overweight
• Overweight or Obesity
• Prediabetic State

Locations

Country	Name	City	State
Latvia	University of Latvia, Faculty of Medicine	Riga

Sponsors (10)

Lead Sponsor	Collaborator
Jelizaveta Sokolovska	CheckHealth, Euronet-Consulting, HK3 Lab S.R.L., Italy, Medical University of Graz, Scuola di Robotica, Spindox Labs, The Institute of Electronics and Computer Sciences, Latvia, The Italian Liver Foundation, The National Research Council, Italy

Country where clinical trial is conducted

Latvia, 

References & Publications (26)

Al-Shamsi S, Govender RD, King J. External validation and clinical usefulness of three commonly used cardiovascular risk prediction scores in an Emirati population: a retrospective longitudinal cohort study. BMJ Open. 2020 Oct 28;10(10):e040680. doi: 10.1136/bmjopen-2020-040680. — View Citation

Almeda-Valdes P, Cuevas-Ramos D, Aguilar-Salinas CA. Metabolic syndrome and non-alcoholic fatty liver disease. Ann Hepatol. 2009;8 Suppl 1:S18-24. — View Citation

American Diabetes Association. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2021. Diabetes Care. 2021 Jan;44(Suppl 1):S15-S33. doi: 10.2337/dc21-S002. Erratum In: Diabetes Care. 2021 Sep;44(9):2182. — View Citation

Bernabe-Ortiz A, Perel P, Miranda JJ, Smeeth L. Diagnostic accuracy of the Finnish Diabetes Risk Score (FINDRISC) for undiagnosed T2DM in Peruvian population. Prim Care Diabetes. 2018 Dec;12(6):517-525. doi: 10.1016/j.pcd.2018.07.015. Epub 2018 Aug 18. — View Citation

Bleeker SE, Moll HA, Steyerberg EW, Donders AR, Derksen-Lubsen G, Grobbee DE, Moons KG. External validation is necessary in prediction research: a clinical example. J Clin Epidemiol. 2003 Sep;56(9):826-32. doi: 10.1016/s0895-4356(03)00207-5. — View Citation

Castiglione F, Tieri P, De Graaf A, Franceschi C, Lio P, Van Ommen B, Mazza C, Tuchel A, Bernaschi M, Samson C, Colombo T, Castellani GC, Capri M, Garagnani P, Salvioli S, Nguyen VA, Bobeldijk-Pastorova I, Krishnan S, Cappozzo A, Sacchetti M, Morettini M, Ernst M. The onset of type 2 diabetes: proposal for a multi-scale model. JMIR Res Protoc. 2013 Oct 31;2(2):e44. doi: 10.2196/resprot.2854. — View Citation

Centers for Disease Control and Prevention. National Diabetes Statistics Report 2020 Website. https://www.cdc.gov/diabetes/pdfs/data/statistics/national-diabetes-statistics-report.pdf

Ellahham S. Artificial Intelligence: The Future for Diabetes Care. Am J Med. 2020 Aug;133(8):895-900. doi: 10.1016/j.amjmed.2020.03.033. Epub 2020 Apr 20. — View Citation

Ezzati M, Riboli E. Can noncommunicable diseases be prevented? Lessons from studies of populations and individuals. Science. 2012 Sep 21;337(6101):1482-7. doi: 10.1126/science.1227001. — View Citation

Hegde H, Shimpi N, Panny A, Glurich I, Christie P, Acharya A. Development of non-invasive diabetes risk prediction models as decision support tools designed for application in the dental clinical environment. Inform Med Unlocked. 2019;17:100254. doi: 10.1016/j.imu.2019.100254. Epub 2019 Oct 16. — View Citation

International Diabetes Federation. IDF Diabetes Atlas, 10th Edn. Brussels, Belgium: 2021. Available at: Https://www.Diabetesatlas.Org.

Jolle A, Midthjell K, Holmen J, Carlsen SM, Tuomilehto J, Bjorngaard JH, Asvold BO. Validity of the FINDRISC as a prediction tool for diabetes in a contemporary Norwegian population: a 10-year follow-up of the HUNT study. BMJ Open Diabetes Res Care. 2019 Nov 28;7(1):e000769. doi: 10.1136/bmjdrc-2019-000769. eCollection 2019. — View Citation

Karniadakis GE, Kevrekidis IG, Lu L, Perdikaris P, Wang S, Yang L. Physics-informed machine learning. Nat Rev Phys. 2021;3(6):422-440. doi:10.1038/s42254-021-00314-5

Ley SH, Schulze MB, Hivert MF, Meigs JB, Hu FB. Risk Factors for Type 2 Diabetes. In: Cowie CC, Casagrande SS, Menke A, Cissell MA, Eberhardt MS, Meigs JB, Gregg EW, Knowler WC, Barrett-Connor E, Becker DJ, Brancati FL, Boyko EJ, Herman WH, Howard BV, Narayan KMV, Rewers M, Fradkin JE, editors. Diabetes in America. 3rd edition. Bethesda (MD): National Institute of Diabetes and Digestive and Kidney Diseases (US); 2018 Aug. CHAPTER 13. Available from http://www.ncbi.nlm.nih.gov/books/NBK567966/ — View Citation

Palumbo MC, de Graaf AA, Morettini M, Tieri P, Krishnan S, Castiglione F. A computational model of the effects of macronutrients absorption and physical exercise on hormonal regulation and metabolic homeostasis. Comput Biol Med. 2023 Sep;163:107158. doi: 10.1016/j.compbiomed.2023.107158. Epub 2023 Jun 16. — View Citation

Palumbo MC, Morettini M, Tieri P, Diele F, Sacchetti M, Castiglione F. Personalizing physical exercise in a computational model of fuel homeostasis. PLoS Comput Biol. 2018 Apr 26;14(4):e1006073. doi: 10.1371/journal.pcbi.1006073. eCollection 2018 Apr. — View Citation

Piovani D, Nikolopoulos GK, Bonovas S. Non-Communicable Diseases: The Invisible Epidemic. J Clin Med. 2022 Oct 8;11(19):5939. doi: 10.3390/jcm11195939. — View Citation

Prana V, Tieri P, Palumbo MC, Mancini E, Castiglione F. Modeling the Effect of High Calorie Diet on the Interplay between Adipose Tissue, Inflammation, and Diabetes. Comput Math Methods Med. 2019 Feb 3;2019:7525834. doi: 10.1155/2019/7525834. eCollection 2019. — View Citation

Riley RD, Ensor J, Snell KI, Debray TP, Altman DG, Moons KG, Collins GS. External validation of clinical prediction models using big datasets from e-health records or IPD meta-analysis: opportunities and challenges. BMJ. 2016 Jun 22;353:i3140. doi: 10.1136/bmj.i3140. Erratum In: BMJ. 2019 Jun 25;365:l4379. — View Citation

Rodrigues PM, Madeiro JP, Marques JAL. Enhancing Health and Public Health through Machine Learning: Decision Support for Smarter Choices. Bioengineering (Basel). 2023 Jul 2;10(7):792. doi: 10.3390/bioengineering10070792. — View Citation

Stolfi P, Valentini I, Palumbo MC, Tieri P, Grignolio A, Castiglione F. Potential predictors of type-2 diabetes risk: machine learning, synthetic data and wearable health devices. BMC Bioinformatics. 2020 Dec 14;21(Suppl 17):508. doi: 10.1186/s12859-020-03763-4. — View Citation

Tabak AG, Herder C, Rathmann W, Brunner EJ, Kivimaki M. Prediabetes: a high-risk state for diabetes development. Lancet. 2012 Jun 16;379(9833):2279-90. doi: 10.1016/S0140-6736(12)60283-9. Epub 2012 Jun 9. — View Citation

Tsalamandris S, Antonopoulos AS, Oikonomou E, Papamikroulis GA, Vogiatzi G, Papaioannou S, Deftereos S, Tousoulis D. The Role of Inflammation in Diabetes: Current Concepts and Future Perspectives. Eur Cardiol. 2019 Apr;14(1):50-59. doi: 10.15420/ecr.2018.33.1. — View Citation

Yang J, Qian F, Chavarro JE, Ley SH, Tobias DK, Yeung E, Hinkle SN, Bao W, Li M, Liu A, Mills JL, Sun Q, Willett WC, Hu FB, Zhang C. Modifiable risk factors and long term risk of type 2 diabetes among individuals with a history of gestational diabetes mellitus: prospective cohort study. BMJ. 2022 Sep 21;378:e070312. doi: 10.1136/bmj-2022-070312. — View Citation

Yao H, Shum AJ, Cowan M, Lahdesmaki I, Parviz BA. A contact lens with embedded sensor for monitoring tear glucose level. Biosens Bioelectron. 2011 Mar 15;26(7):3290-6. doi: 10.1016/j.bios.2010.12.042. Epub 2010 Dec 31. — View Citation

Zafar H, Channa A, Jeoti V, Stojanovic GM. Comprehensive Review on Wearable Sweat-Glucose Sensors for Continuous Glucose Monitoring. Sensors (Basel). 2022 Jan 14;22(2):638. doi: 10.3390/s22020638. — View Citation

* Note: There are 26 references in all — Click here to view all references

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Validation of the Mission T2D (MT2D) algorithm outputs, that predicts the real time risk for developing pre-diabetes.	Data collections has three main purposes input data for the in-silico MT2D model (gender, weight, height, number of sessions of physical activity, duration of the bout of physical activity, intensity in terms of %VO2max, 3 meals per day (specified macronutrients).; Validation of the MT2D outputs include inflammation markers, metabolic outcomes. The third data for training/validation of the physics-informed machine learning (PIML) algorithm: demographic data; health-related data; lifestyle data (e.g., food consumption data and physical activity data); continuous ingestion through wearable sensors (Continuous Glucose Monitoring (CGM and tracker of physical activity e.g., Fitbit Charge 5, EDIBit.)	The study will run for 15 months. During this period, 75 individuals will be followed for 4 months, including screening visit and three onsite visits, if participants meet the predetermined inclusion criteria. Time frame between visits are 65 days (± 10