IntelliHearts

As we can see in the Pandemic situation, home remote monitoring and detection of vital signs is really important. We focused on who as Cardiac diseases (first cause death in the world -  around 31%) and who as respiratory diseases (+250 mln with COPD). We are able to run a continuous health and vital parameters tracking and AI analysis from smartphone paired with one of the compatible devices. It could potentially help who is affected, monitoring, and do prevention with early discoverning. 

home remote monitoring and detection of vital signs is really important. We focused on who as Cardiac diseases (first cause death in the world -  around 31%) and who as respiratory diseases (+250 mln with COPD) - OMS stats - 

We had the first version of the App and the first device integrate.
ALL OUR ALGORITHM USE THE INTER-PATIENTS SCHEME DIVISION, THE ONLY ONE THAT IS RECOGNIZED IN MEDICAL DETECTION, this is really important when we talk about ML in medical field.
We had ML/AI algorithm that outperforms the state of the art using medical separation scheme:
95% on heartbeat classification for arrhythmias and atrial fibrillation detection
95% on audio respiratory disease detection.
On ML audio algorithm is totally new and invented by us. For technicians our Deep Net doesn’t turn audio in image and analyze it. It works directly on raw audio, we will publish a paper about that. It’s computationally more lightweight bringing it inside the app.
Using existing hardware paired with our app we are able to analyze the signals and track other vital signs such as oxygen saturation, blood pressure, heart rate, sleep tracking, HRV, apnea.

The target is mainly people between the ages of 30 and 60. We understand them because the mother of the CTO has a stroke (he starts building this device for himself) and the CEO's mom has had cancer and she would like to have this type of device to monitor.
In addition, we are in contact with a few hospitals to start a prototype trial and will be tested by doctors and patients.

Because is based on prevention and in Healthcare is the most important part. Plus using ML algorithm on servers we could continue to improve feature just doing new updates. 

The technology is the result of years of research in the field of pattern recognition and deep learning with the aim of developing real-time computer-aided diagnosis tools for early disease detection and prevention. Each part of the technology such as arrhythmias heartbeat classification, breath audio classification, were or are currently being published in major scientific journals and conferences.
The main competitor will be Apple watch and Kardia but our devices are on a completely different level, we are talking of less than 100 euros which are at least one-fifth of the price of an Apple watch. We are more into a health smart band with features missing in Apple watch and Kardia such as oxygen saturation monitoring and blood pressure.
None of them has an Audio respiratory algorithm that could be important especially chained to Covid-19. We are waiting for the release of the Dataset of Cambridge university to use our algorithm. We are pretty sure that we could outperform the results. 

Scientifically proven Machine Learning and AI algorithms: 

Here we attach the 2017 position paper.
Essentially the technology used is that presented in that paper. From a technical point of view, the deep learning neural network used in that work was replaced with the newer ResNet-152 giving a leap of accuracy of 93% in F1-Weighted score for arrhythmias classification and 95% F1-Weighted in atrial fibrillations (not yet described in that paper).
With IntelliHearts we mainly use sensors related to ppg and ECG, and we are able to detect:
Bradycardia, tachycardia, fusion beats and ventricular and supraventricular ectopic beats.

In the experiment that was conducted we first acquired the signal and cleaned it with statistical techniques (e.g. wavelet) then inter-patient separation was made. Subsequently every single beat is plotted and shown in 2d centered in the peak R.
Consequently we have n beats that can fall into the categories: normal, ectopic, supraventricular and fusion. There are not always the same number of beats within the same class, for this reason we refer to the F1 microscore average.

Respiratory diseases detection was conducted in the following way:

• We used the data from Paper: Α Respiratory Sound Database for the Development of Automated Classification
• Filter noise and remove background sounds
• We used several segmentation techniques based on spectrograms of respiratory audio
• We used a custom deep learning neural network to predict binary healthy/unhealthy patients using only audio recorded by smartphone

Demo ECG and AI detection
https://www.youtube.com/watch?v=3D_TLbO8i5g
Whit the audio respiratory diseases we were finalist into a hackathon organized by IEEE (Quarantine Hackathon https://devpost.com/software/a...)

I don't need too many words. If everyone can track his health is easier to do prevention. Prevention is stronger that cure EVERY TIME. 
So it is better to spend 100€ on a device/platform that monitors vital signs, do AI elaboration of ECG and give to you the opportunity to find problems in time or wait that your body will say you "Hey, we have a problem" or wait and spend more and more money in cure?.I mention again cardiac disease is the first die reason in the world

2 full time

Because this is not the first Startup for either founder, and we faced both failure and success in our past, and as an honorable mention, the CTO is the CTO and founder of Nextome SRL, that is a successful startup that generated over 2mln euros only in the last year, and winner of SME and international competitions. 
Plus on the technical side, the CTO has an MSc in AI at Georgia Tech (USA) and is a Ph.D. student as machine learning in pattern recognition. 
On the entrepreneurial side, the CEO has years of experience in entrepreneurship in his family business, in production and export.
We can bring this project to the next level because we have experience, creativity, vision, parsimony (we don’t like spend money that are not needed, e.g. we are running all just using deals that we found and got) and because thanks to our skills. Intellihearts is now running on a $5 server with the state of the art for audio respiratory diseases on! Also, our entire network and the presence on newspapers was obtained free of charge.

1) Find companies that want resell our products. IntelliHearts wants import and provide hardware and software. We are looking for companies that are interested to sell our hardware buying it in stock.
2) Find companies that have connections with Hospitals and have clients in Healthcare. We are trying to see if we can use them as “trojan”, most of this kind of company that can’t provide high level AI algorithm but need it to acquire more interest, so partnering with them will be a win-win solution.
3) Sell AI API to third parties