PreHypo: AI-Powered Service for Advanced Prediction

PreHypo is an innovative solution for CGM manufacturers looking to improve their devices with AI-driven prediction. By combining advanced machine learning with CGM data and biomarkers such as heart rate, we can revolutionize the diabetes management market and give manufacturers a competitive advantage. PreHypo is patented and clinically validated through extensive studies and millions of data points, proving its accuracy and effectiveness.



CASE: We help predicting low blood sugar in due time

Hypoglycemia is a feared condition for patients with both type 1 and type 2 diabetes since it can result in brain damage, coma and in worst case even death. Hypoglycemia can occur if you eat less than usual after taking your regular dose of diabetes medication, or if you exercise more than you typically do. Our invention enables patient empowerment by triggering a customize and personalized alarm from the patients or relative’s smartphone / smartwatch. This enables early intervention to avoid or reduce hypoglycemia.




What is Hypoglycemia



A condition where blood sugar falls below a critical level and disturb organ functions. It is often because of too much insulin combined with too little food intake. Low blood sugar (hypoglycemia) is the most common and most dangerous condition for many people with type 1 diabetes (T1D). Very low blood sugar may lead to insulin shock, which can be life threatening if not promptly treated. Recent reports have indicated that the incidence of hypoglycemia requiring emergency assistance is as high as 7.1% per year and that as many as 6-10% of all deaths in patients with T1D result from hypoglycemia. Patients with T1D are known to have increased risk of sudden, unexpected death during sleep. This phenomenon, known as “dead-in-bed syndrome,” occurs due to mechanisms that are poorly understood.

...









Technology

Using the latest advancements in AI algorithm development, our invention enables accurate hypoglycemia prediction for people with diabetes. Emerging technologies like Continuous Glucose Monitoring (CGM) have revolutionized real-time glucose tracking, allowing users to monitor trends and potentially act before reaching critical levels. However, CGM readings can sometimes lack the precision and reliability needed to detect impending low glucose events. MEDICUS Engineering’s clinically validated and patented cloud-based software algorithm, PreHypo™, enhances CGM technology by significantly improving prediction accuracy, sensitivity, and lead time. By leveraging cloud computing, PreHypo™ ensures seamless, real-time data processing and accessibility, optimizing CGM data to deliver timely, actionable insights. This innovation empowers users and healthcare providers to intervene proactively, increasing the chances of preventing hypoglycemic episodes before they occur.

PrePreHypo™ is a cloud-based software algorithm that seamlessly integrates into existing CGM systems, smartphones, smartwatches, fitness trackers and digital health management platforms. Its flexible installation and cloud connectivity enable real-time data processing and predictive insights, enhancing the functionality of current diabetes technologies. By integrating PreHypo™, users gain access to advanced prediction, empowering proactive management and timely intervention.



40 minutes prediction time

Our sophisticated mathematical models predicts hypoglycemia up to 40 minutes before it happens.

Patented technology

Our intellectual property is backed with new patented technology and equipment.

Validated

Validated and tested on 18.5 million minutes of CGM and 57 million minutes of synthetic CGM data.

Performance

Performance has been shown to yield an ROC AUC of 0.98




Publications

We have proven the power of our technology which has been tested in both clinical setting and during daily living. We are currently setting up new clinical trials to test and further validate our technology.

Hypoglycemia event prediction

Front. Clin. Diabetes Healthc., 09 December 2022.

Spontaneous hypoglycemia

J Diabetes Sci Technol. 2014 Jul;8(4):731-7. doi: 10.1177/1932296814528838.

Predicting hypoglycemia

J Diabetes Sci Technol. 2015 Jan;9(1):132-7. doi: 10.1177/1932296814549830

Validation

J Diabetes Sci Technol. 2019 Nov;13(6):1178-1179. doi: 10.1177/1932296819864625

Hypoglycemia

Diabetes Technol Ther . 2017 Feb;19(2):91-95. doi: 10.1089/dia.2016.0342.

Get in Touch

We actively welcome new partnership opportunities

MEDICUS engineering ApS
Palle Juul-Jensens Boulevard 82
Denmark, 8200 Aarhus N
mail@medicuseng.com
NEWS
MEDICUSeng.com