“The Decentralized AI Alliance: The Proof of Fusion that will give a further boost to Medicine and Healthcare.”

graphics source: @theodysseyonline

The 21st century is the information age; human beings are hungrier for information and, in particular, they need information related to their health.

The source of the information is the data. Data from electronic medical records, clinical trials, genome, mobile app, wearables devices and the Internet of Things (IoT). When we talk about AI applied to the field of medicine, research and healthcare, we talk about this exactly: bigdata.

Today patients can proactively manage their health and contribute with their personal data to predictive medicine, clinical research, precision medicine, drug discovery, research on aging, telemedicine, traceability of counterfeit drugs and more.

Our medicine has never been as advanced as it is today. We have robotic surgery, cell therapy, DNA sequencing and genetic engineering. However, there’s a surprising paradigm: our health system is at the same stage as it was during the Stone Age: because it is curative and based on treatments.

In most cases, the starting point is the visibility of the disease.

We wait that health is degraded enough and that symptoms manifest, revealing the presence of the disease. Great part of the population does not want to deal with healthcare while feeling healthy. Only when they feel ill, they interact with the problem.

A healthcare system, based on symptoms, creates a too long and dangerous period between the beginning of the disease and the start of the treatment and this involves complications, both in terms of health and in terms of treatment costs.

So, knowing the limits of this curative medicine system, it is necessary to move towards a health system based on prevention and the key factor in this sense is represented at the moment by decentralized artificial intelligence.


The stakeholders of the health system are represented by medical professionals, pharmaceutical companies, insurers and governments, who agree on the value and potential of preventive healthcare but do not carry out any coordinated action

The central point, however, is another: as said before, most of the population interact with the doctors only when they feel bad. This means that most of the data collected comes from sick people and this devalues all data for the purpose of prevention.

The solution is empowering people to collect data.


The problems that characterize the field of medicine are varied but fundamentally they concern the large amount of misdiagnoses and missed diagnoses, especially those related to rare diseases. Doctors lack the time to analyze medical history and documentation, something that the AI can easily overcome, and in addition AI is not susceptible to sleep deprivation, distraction, and information overload.

Artificial intelligence is transforming all the person’s data into medical data, so we need a secure and transparent distributed data market that uses blockchain and deep learning to solve the challenges faced by medicine, but also returns control over personal data to patients. The information provided by users and necessary for machine learning to develop new knowledge should not be entrusted to a centralized entity but remain distributed.


Biomedical data can be recovered from biomedical imaging, laboratory tests based on blood tests, genetic profiling and DNA testing. Costs of genome analysis will become in next years really cheaper and a lot of people will do this kind of analysis. This is expected: that the amount of genomic data will exceed the amount of data generated by other fields like for example social networks.

What we need to focus on is that the role of data is constantly present during the patient’s life. In congenital genetic diseases, the records obtained in the first years of life are important for determining the development of the disease. For the same reason it is important to analyze the results obtained before the diagnosis of illness and monitor our health data always and constantly.

Data subjected to AI analysis can be divided into the following categories:

dynamic: reflecting the state of the organism at the time of sampling (blood test, transcriptome, epigenome, proteome, microbiome etc.); static: almost unchanged during the patient’s life (genome, fingerprint);

The analysis of these two kind of data is fundamental: patients with the same diagnosis can respond to the same therapies in different ways, a drug, for example, can be particularly effective for a patient, while for other patients, the same drug may not respond in the same way. Decentralized Artificial Intelligence and Big-Data analysis platforms for access to anonymized data in conjunction with bio-informatic and bio-statistical models and clinical studies will give rise, to a medicine of precision, tailor-made for the patient.


Differently from what most people imagine, the impulse to research on Artificial Intelligence did not come from the computer field but from the biological one. In 1969, in fact, some students and researchers of the Carnegie Institute of Technology created a program, called DENDRAL, which helped chemists to identify unknown organic molecules, analyzing their mass spectra and using chemistry knowledge.

After many years, in 2013, the European Commission launches the Human Brain Project to realize a complete simulation of the human brain by 2023.

In the same year, IBM announces the first commercial application of Watson in oncology at the Memorial Sloan-Kettering cancer center. How does it work: when a doctor asks a question specifying the symptoms and other related factors, Watson first processes the information to identify the most important parts, then processes the patient’s data to find facts relevant in the patient’s medical and also hereditary history, then examines the data available from the sources to formulate and test hypotheses, and finally provides a list of individualized recommendations classified by level of evidence. The data sources include treatment guidelines, electronic medical records, doctor and nurse records, research materials, clinical studies, journal articles and patient information. At the moment Watson is a real decision-making assistant for medical staff.

In 2016, Google’s DeepMind Health project takes off to improve eye care and kidney damage and breast cancer diagnosis. Originally it has been acquired by Google in 2014 to create simulation programs and online games and now it also working in the field of predictive medicine. Google researchers are able to detect through AI the spread of breast cancer in lymph node tissue on images of microscopic samples with an accuracy comparable to (or greater than) human pathologists.


Among the examples of AI application in medicine there’s:

the recognition of the vocal pattern for the diagnosis of Parkinson’s disease and its gravity prediction; it has been realized also an app for smartphones that uses the gyroscope in many mobile devices to analyze and quantify tremors, gait and performance in a “finger taping” test. An AI algorithm distinguishes between actual tremors and “incorrect data”. This tool allows patients to perform a home test, providing valuable and quantitative feedback on how their personal lifestyle and medications can affect their symptoms;

personalized diet to avoid glucose spikes after eating related to the specific foods we take and our intestinal microbiomes;

the diagnosis of diseases such as tuberculosis (TB) from chest x-ray images, eye diseases and cardiological diseases;

the monitoring and control Of Asthma: an APP was created, Asthma MD, which coupled with a portable flow meter that measures pulmonary performance by evaluating peak flow during expiration, records data for people with asthma but also other respiratory diseases. Users can subscribe to a program where their data is uploaded anonymously to a Google database that is assembled for research purposes. Anonymous data will help correlate asthma with environmental factors, triggers and climate change.


The Blockchain and the AI, actually, come into play in the healthcare sector also in the detection of counterfeit drugs, increasing the safety and transparency of the pharmaceutical industry and tracing the pharmaceutical industry globally. The AI will detect the counterfeit drugs along the analysis of the entire supply chain and will automatically report the data that will be tracked on Blockchain. But the fight against counterfeiting will cover not only drugs but also prostheses, pacemakers, surgical instruments, avoiding sources of error concerning compatibility as well as counterfeit-related scams.

With nearly 80% of clinical studies failing to meet enrolment times, a more efficient approach is essential that allows real-world data to be used to identify areas and clinical centres with the highest patient availability potentially suitable for study. Through communication on social media it is then possible to attract patients who wish to experiment with new therapies and collect voluntary compliance with the study through electronic informed consent. Subsequently, through remote monitoring applications, the patient can record in real life, a series of parameters of his clinical condition during the study, remaining quietly at home, without having to repeatedly go to the centre for checks and thus saving time and money.


AI is still limited by the quality of the data it can access. As a result, blockchain has emerged as a core asset for healthcare IT. Currently, healthcare blockchain is mostly financially focused (insurance and monetization of data) while we have to work to make blockchain effective for other aspects of healthcare. Blockchain is a necessary element of trust in the AI. In a certain sense it is like Blockchain itself was a sort of form of AI: a code that will run by itself, improve itself and get smarter and harder, it will require more energy and resources, this code already exists: it is called blockchain.

Blockchain can be used to publish metadata about existing data in a consortium of health organizations. These metadata can include pointers to business systems that store data and hashcodes that can be used to verify data integrity. The organizations participating in this blockchain can discover the available data, identify them and then request the data of interest through a secure and direct peer-to-peer exchange. Blockchain can protect both the data stored on the blocks of the chain and the data stored off-chain and referenced by metadata, pointers and hashcode.


It’s interesting to consider that the value of various data types may depend on the application. For example, for insurance companies, while the cost of generating data can be significantly higher for the genome than for a selfie, the value of the recent patient image can significantly exceed the value of the genome, since it may be more predictive than the age of the patient, health status and mortality. However, combining these types of data will be significantly more valuable than the value of these data types individually.


Medical data is a valuable asset and any asset that has value can be considered a currency. People who generate data should own that currency and be able to share, sell, buy, or do whatever else they choose to do with it. We have to give to user full ownership and control over their medical data: to can easily control how, when, and with whom their data is shared.


If you generate data, you should own it. If they have value, you should have a part of it. At the moment big data is oligopolized by the big companies: Facebook, Google, Amazon. The decentralized AI could give people back the full ownership of this data, reduce the monopolistic power of technological giants and allow citizens to share data with anyone on their terms to improve their well-being and health. This allows data markets without intermediaries, companies and researchers will be able to search for relevant information without having to know the identity of the data owners and this represents the enormous impact of being able to filter and identify the desired data while keeping the users’ privacy intact.


In health care, healthbots are used in multiple ways: in booking appointments, in checking the identity of patients, in insurance and coverage information, in requesting medical history or in taking medication and supplying information on side effects and drug interactions.

The first bot “operating” in the health world seems to date back to about fifty years ago: it was called ELIZA and was created to imitate a Rogerian psychologist: a therapist who asks questions to the patient simply re-elaborating what the patient himself said; Currently, chatbots like HealthTap and Your.Md are operative: they monitor symptoms, ask for information and help to alleviate them; the application is based on an advanced AI algorithm.

In case of therapy to be followed, then, there is another useful chatbot that can help you remember to take prescribed drugs: it’s called Florence and it’s a real “Personal Nurse”. We can tell the bot the name of the medicines we have to take, how many times a day and at what times, also setting the duration of the therapy.

The real usefulness of this type of chatbot is, currently, the possibility of providing advice and information for a healthy life.

Many young people have no sex education or knowledge about sexually transmitted diseases, because it is considered a taboo subject in the family, at school and in the community.

Most of the population does not know the correct use of drugs like antibiotics.

The ambition is to create a decentralized chatbot network, able to respond quickly to questions in a relevant way and with the highest standards of data protection and privacy.


Doc.ai is based on the concept of connecting people, researchers, and companies to help accelerate medical research in an engaging framework that respects people’s ownership of their data and rewards them with their fair share has been the core value of our team from day one.

Users opt-in to share specific data with the doc.ai platform on a specific research project. The platform enables people to participate in medical research by sharing their real-world data to accelerate research through the application of AI technology. This enables researchers to build predictive models using the data they have shared for research to be turned into actionable knowledge to benefit humanity.

It’s important to note that the shared data can only be used for the project a user agreed to join and help.

When you participate in the research data trials on doc.ai, you get rewarded with points in your wallet and you can exchange these points on the doc.ai marketplace for Amazon gift cards. These transactions are designed to use smart contracts.

Skychain is a data marketplace that allows healthcare professionals around the world access neural networks, paying the developers a fee for each use of their neural networks. Doctors can do preditcion analysis for: Respiratory Diseases Recognition, Bone Abnormalities, Skin Cancer Detection, Breast Cancer Detection, Liver Cancer Detection, Brain Disorders Detection.


Privacy has been recognized as a fundamental human right by the United Nations in the Universal Declaration of Human Rights in 1948, and there is still no universal agreement on what constitutes privacy.

Now it is really difficult for third parties to monetize data without the clear consent of their rightful owners. With the development of regulation about the abuse of data management like GDPR (May 2018), the healthcare industry is now shifting its operations to comply and use people’s data with their full and explicit consent. This advancement of everyone’s fundamental rights shows that individuals are not ready to give up their privacy.

More specifically, the regulation targets patient consent and patient data. The law requires companies to de-identify and anonymize any clinical data before it is analysed. The types of data considered personal under GDPR include:

• Identification data: (name, ID number, address, email, social media accounts, etc.).

• Personal data: physical, genetic, mental, economic, cultural or social identity.

• Genetic data: Personal data relating to the inherited or acquired genetic characteristics

• Data concerning health: Personal data related to an individual’s physical or mental health, including provision of health care services.

• Biometric data: Personal data resulting from specific technical processing (i.e. facial recognition, fingerprint analysis)

GDPR’s data privacy rules can be particularly challenging when healthcare companies are working with a curated collection of data sets or reviewing data about specific populations. For example, if there are only a handful of patients living with a certain condition, prescription data combined with electronic case report forms could easily expose their identity.

Generating high-value data and insights within the framework of GDPR is possible. It requires data science teams with technical, regulatory, and pharmaceutical industry expertise to stand up sophisticated processes for data management.


The problem in essence with artificial intelligence is that it remains a centralized solution in which data is stored centrally, owned and controlled by the group that collects them. AI alone cannot radically change the current business model in the health sector because it does not have the capacity to create a decentralized and participatory economy of shared knowledge and insights.

The blockchain solves the problem of protecting users’ privacy, data distribution, protection of data sets from bias, manipulation and hacking and offers a guarantee of data transparency. A fully decentralized and tokenized data exchange system with a transparent reward model will be resistant to attacks, collusion and censorship.

Healthcare is a, fragile and misaligned ecosystem that lacks trust. Therefore it needs decentralized AI to transform patients from healthcare consumers into health producers.

Healthcare organizations are in the business of providing care for patients — not building and managing information technology — just IT support and emergent technologies (Blockchain, AI, IoT) will make the healthcare system change.

#Blockchain Ladies Founder, #ICO Advisor & Community Manager. Digital #Health Expert & Neuroscientist. @BchainLadies @katerinaFerrara