NURSY

Grupo Latinoamericano de Enfermedades Cronicas (GLEC)

10% of people affected by rare diseases are in LATAM.

The fourth most common RD globally is cystic fibrosis, currently our main focus in Argentina.

The evidence also demonstrates the economic burden of RDs due to high direct and indirect costs on health systems, and the leading categories for direct costs include inpatient and outpatient services, RD expenditure in the USA was estimated to be USD 966 billion in 2019, from which USD 418 billion corresponded to direct medical costs and USD 548 billion to indirect and non-medical costs. Still, these values are higher than those estimated for some well-known chronic diseases. But also the financial burden incurred due to out-of-pocket RD treatment spending might put patients and their families in economic vulnerability.

NURSY is a virtual assistant driven by AI, she is online 7x24 and can hear and talk to people, using any technology whether texting by any chat IRC application such as Telegram, WhatsApp, WeChat, etc, or SMS

NURSY can hear people's problems and detect hazardous situations using AI-driven (medically trained) triaged questionnaires performed in real time. Then as soon as possible (ASAP) NURSY can contact a medical team on the other side, in a day-night shift at any health prevention center, and make a resume of the patient's situation, taking note of the indications, and tell them to the involved patient.

NURSY is not only able to react, but she can act proactively, by checking by med-protocol on a timely basis for example if the doctor prescribed antibiotics, and some aspirin, for the fever. NURSY gets the indications from the doctor, (on his terminal or mobile phone) and performs the tasks just like a loving mom or dad, giving the correct indication in time, and asking for the fulfillment, just like: "Hey, did you take the pills at noon?" and if the patient does not answer, she will insist until a deadline, and in case of a serious sickness, she can report back to the treatment medic team and take the necessary actions.

NURSY can save even lives, when the situation is bad, and solve urgent calls because she can undertake a "normal" human conversation using her internal brain" (an advanced health-targetted AI system), which is tuned to detect severity and specific situations.

NURSY is rather useful for normal people going to a doctor, but she is especially useful for endangered patients, as well as for rare or chronic diseases, people with memory problems just as elderly people, and disaster situations where there is no electricity nowadays phones shown us can survive most situations and be useful for saving lives.

DEMO VIDEO: https://api.adipta.co/content/...

Our target population is the Cystic Fibrosis patient population in the Hospital Municipal de Rehabilitación Respiratoria María Ferrer located in Buenos Aires City, Argentina.

Like all patients of traditional medicine, each of them usually comes into contact with their doctor every 2 to 3 months, but in the meantime, things happen to them.

We believe without fear of being wrong that for the first time, we are managing to avoid the dangerous exacerbations that Cystic Fibrosis can cause death to a patient.

Because NURSY simulates the behavior that a specialist doctor would follow if he could talk to his patients frequently and review the 5 most important variables such as fatigue, lack of appetite, cough, expectorations, and dyspnea.

In this way, NURSY can visualize a possible exacerbation before it happens, managing to warn the Doctor as if he had been talking to his patient so that he can make the best decision and tell NURSY what to tell him. from now on.

Then, NURSY, following the doctor's instructions, informs the patient of what is indicated and begins to follow up to corroborate that the patient is doing the task, managing to inform the doctor of the progress or problems that the patient again indicates.

We observed that the patient population registered an 80% adherence to the solution, in other words, patients responded to NURSY interactions in 80 out of 100 interactions on average.

NURSY is especially useful for endangered patients and populations with a poor cultural level where people do not realize the danger of not fulfilling a treatment or not being able to detect a dangerous situation of their health status.

NURSY is specially intended for rare and chronic diseases, as well as people with memory problems just as elderly people.

NURSY can be a dramatic help for populations in disaster situations where there is no electricity or water, food, authorities, or medical aid, but nowadays, cell phones have shown us they can survive most situations and be useful for saving lives, being there and connecting people with medical aids, the problem we solve is concurrency under disaster situations, NURSY can multiplex herself into as many users as needed by scaling dynamically, providing help to thousands of persona at the same time, in real-time.

NURSY helps also the medical system, bettering treatment efficiency, lowering costs by avoiding exacerbations by prevention, and reminding people to perform their medical duties in time, every day, at every needed moment, just as Mom or Dad did with us when we were kids.

And even though NURSY knows only about Cystic Fibrosis today, it is a matter of giving it knowledge about an additional set of rare diseases so that it can help many more patients.

We built a NURSY solution focused on Cystic Fibrosis (CF), backed by a prominent specialist Doctor named Oscar Rizzo, who has opted for innovation as a way to help his patients and works hand in hand with us to make NURSY the best possible solution.

With Oscar's support, we implemented the solution in a population of 25 patients in the Hospital Municipal de Rehabilitación Respiratoria María Ferrer located in Buenos Aires City, Argentina.

NURSY can interact with people in a free and open way as human beings used to, we comment, say, or express something and the other human being can understand me, that's the way NURSY interacts.

NURSY can contain the patient's anxiety when they want to express things that happened to them

It can also help patients better understand and educate themselves about their disease.

But also for the first time, it allows the Doctor to know where he should focus his time and resources, indicating those patients who have a yellow or red condition according to how he has instructed NURSY to notify him.

NURSY will never perform a medical act nor seek to invent an answer for a patient, its communication will always be supported by the knowledge of a doctor who has been indicated to it.

All the conversational data of the interactions is the property of the medical institution that uses it, achieving its analysis to facilitate collaborative research and accelerate new treatments.

Finally, NURSY saves costs in the medical field because it allows doctors' efforts to be aligned towards those patients most in need, and by predicting exacerbations, it avoids patient crises that cause hospitalizations and high costs for the health system.

GLEC is a not-for-profit organization working so hard to expand the knowledge about our solutions, we consider that the Amgen Prize could help us to get the recognition that we need to broaden our solution.

The cash prize would allow us to expand the spectrum of our solution to a universe of 500 people, which would allow us to move to a Growth phase with an established solution available in more communities and with a consistent design and approach.

The same concept of NURSY allows us to achieve a solution meaningfully guided by the communities' input, ideas, and agendas because being a free and open conversational solution allows patients or even their caregivers to express opinions, ideas, suggestions, or complaints that may be channeled appropriately either to improve the solution or as learning about what the market is demanding at all times, and observe how that feedback may be changing at all times.

It is that ability to express oneself freely where NURSY allows us to collect relevant information that Improves data standardization, centralizing, and sharing to develop mechanisms to support the identification of RDs and collectively develop new ways of defining and measuring value.

For this, we have a Data Analyst who allows us to identify trends within the conversational flow that develops with patients.

Assuming 1 en 10 people affected by Rare Disease

Current 2024 LATAM Population: 670 M 

IMPACT GOALS 2025

Number of RDs known by NURSY: 2

Patients connected with NURSY: 500

Adherence to NURSY: 83%

IMPACT GOALS 2026

Number of RDs known by NURSY: 6

Patients connected with NURSY: 30K

Adherence to NURSY: 85%

IMPACT GOALS 2027

Number of RDs known by NURSY: 15

Patients connected with NURSY: 100K

Adherence to NURSY: 87%

IMPACT GOALS 2028

Number of RDs known by NURSY: 45

Patients connected with NURSY: 800K

Adherence to NURSY: 88%

IMPACT GOALS 2029

Number of RDs known by NURSY: 150

Patients connected with NURSY: 2M

Adherence to NURSY: 89%

IMPACT GOALS 2030

Number of RDs known by NURSY: 500

Patients connected with NURSY: 5M

Adherence to NURSY: 90%

We have already seen it applied in different use cases and published many related papers. The initial system has been informed to congresses in 2011

NOTE: Most communications and conferences are in Spanish due to the regional LATAM language.

Many publications as well as the PhD Thesis' Andres Hohendahl are freely available on this CV Site (BsC in Electronic Engineering Andres Hohendahl) https://pbox.com.ar/fiuba

[1] Hohendahl, A. T.; Zanutto, B. S.; Wainselboim, A. J. “Desarrollo de un algoritmo para la medición del grado de similitud fonológica entre formas escritas” SLAN2007. X Congreso Latinoamericano de Neuropsicología 2007, Buenos Aires, Argentina

[2] Hohendahl A. T., Zelasco J. F., "Lematizador Morfosintáctico y Semántico Robusto con Flexionador y Estimador Idiomático, usando algoritmos eficientes y compactos para idiomas muy ricos como el español"  XII Congreso Argentino de Ciencias de la Computación, Potrero de los Funes, San Luis, Argentina. CACIC 2006 - ISBN 950-609-050-5

[3] Hohendahl, A. T. "Procesamiento de Lenguaje Natural Robusto" ProLen 2011, Primer Encuentro de Grupos de Investigación sobre Procesamiento del Lenguaje, UBA, Facultad de Filosofía y Letras, del 4-6 de mayo 2011, Biblioteca Nacional, CABA, Argentina. 

[4] Hohendahl, A. T. "Plataforma para Desarrollo de Agentes Inteligentes" ProLen 2011, Primer Encuentro de Grupos de Investigación sobre Procesamiento del Lenguaje, UBA, Facultad de Filosofía y Letras, del 4-6 de mayo 2011, Biblioteca Nacional, CABA, Argentina.

[5] Hohendahl, A. T. | Zelasco, J. F. "Algoritmos eficientes para detección temprana de errores y clasificación idiomática para uso en procesamiento de lenguaje natural y texto" WICC2006 - VIII Workshop de Investigadores en Ciencias de la Computación, Universidad de Morón, 2006, ISBN: 950-9474-35-5*

[6] Sobre la Inteligencia de las máquinas, Revista Entelequia Nº174  (nov-dic 2008)

[7] Que es Procesamiento de Texto Natural (NLP) (díalogo coloquial) 

[8] Agentes Inteligentes (para que sirven y como se comparan con otros sistemas)

[9] Tesis (2014) "Detección y Corrección de Errores de Ortografía"

[10] "Estado del Arte en Corrección Ortográfica - (Revista Coordenadas/COPITEC - agosto 2015)"

[11] "Qué es la Inteligencia Artificial - (COPIME/La Revista Nº38 - agosto 2018)"

Full-time Staff: 5

Part-time Staff: 3

Contractors: 2

We have been working on this solution for over 2 years, developed an AI conversational system in partnership with ADIPTA, Inc., NURSY is NOT just a chatbot, her AI system can be empathetic, understand free text, make comments, and tell jokes, and precisely detect when the situation is bad (healthily speaking). 

She can withstand spelling errors, and understand almost all of the medical vocabulary as well as slang, and other ill-behaved situations. 

José Berbeci has outstanding managing capabilities to drive software integration with processes

Andres Hohendahl is an R&D expert, on Artificial Intelligence, professor & researcher in Natural Language Processing (NLP), 40 years of experience in programming, and he created a modern language called Dialog Description Language that runs inside an NLP framework to accomplish almost all the needed tasks to be considered as a Virtual Human.

We also have been backed by the knowledge of our medical advisor, Dr. Jose Luis Ippolito, who is an expert in rare and chronic diseases.

All work was supported and supervised by Dr. Oscar Rizzo, a prominent and dedicated medical professional specializing in Cystic Fibrosis 

The NURSY business model is similar to the human model of paying for services per hour, in this case, an instance of NURSY is activated for each patient who simultaneously requires its services, at a rate of 1 Dollar per hour.
For example, if at a given time in a medical institution, there are 2 patients at the same time requiring the use of NURSY, then that generates a cost of 2 USD for that hour of service, and so on for each hour of the day depending on the number of NURSY activated per hour.

We identify the following ways to fund our work through NURSY Services:

1.- Medical Institutions looking for the best use of their medical resources

2.- Risk Management area in the Insurance Companies looking to minimize the risk of the patients suffering RDs

3.- Data for Pharma industries to discover patterns in the use of different products for our RDs patients.