Introducing Arachnet

Introduction to Arachnet BlindPhysio,

• Arachnet BlindPhysio is a medical diagnostic tool based on principles of artificial intelligence, machine learning and deep learning.

• The field of medicine that Arachnet BlindPhysio deals with is the field of musculoskeletal disorders.

• User of Arachnet BlindPhysio is a physical therapist.

• Arachnet BlindPhysio is accessible and can be used by visually impaired physical therapists as well.

• Arachnet BlindPhysio is developed by the non-profit company Arachnet Project z.s.

• Arachnet Project's aim is to help and support visually impaired physical therapists to broaden their knowledges and create circumnutates for growing quality of their life

• Arachnet Project is open to all visually and physically impaired experts for cooperation and common success

Arachnet BlindPhysio Workflow

• The therapist obtains information of the problems of therapist's client by the client himself.

• The therapist transforms unprofessional description of the client's problems to the medically correctly defined signs and symptoms.

• The transformation to the correctly defined terms is done with a help of Arachnet BlindPhysio, using a sophisticated method based on mathematical principles.

• After all signs and symptoms are identified they are sent from the therapist's mobile device to Oracle Cloud with Web Server, Relational Database Management System and trained Machine Learning Model.

• The signs and symptoms sent by the therapist are processed by ML Model.

• If the input information is sufficient for determining the diagnosis of the client's disorder or disease:
• The ML Model determines the diagnosis and sends it back to the therapist's mobile device
• based on obtained diagnoses the therapist can ask for additional information regarding the diagnosis, e.g. way of therapy or information on anatomy and physiology

• If the input information is insufficient to make a diagnosis:
• Arachnet BlindPhysio can ask for clarifying questions

The problem Arachnet is trying to solve

• Problems of prevalence of musculoskeletal disorders:
• costs of treatment musculoskeletal diseases that is among the highest cost diseases of modern society
• the shortage of health professionals in general and in the field of musculoskeletal disorders in particular
• long waiting times for an appointment with a doctor
• neglection of possible serious diseases which symptoms can be mimic by musculoskeletal disorders

• Employment of disabled people, i.e. visually impaired people, can overcome the problems of:
• lack of rehabilitation doctors and rehabilitation professionals
• high rate of unemployed disabled people
• problems with real integration of people with serious disabilities (visual impairement) to the mainstream society
• lack of a sense of self-worth and selfconfidence of visually impaired people

Proposed solution

• The application is built on principles of artificial intelligence, machine learning and deep learning

• Use of cloud solution for application development and operation .

• Modular solution of the application, Arachnet BlindPhysio is built of several independent modules connected by standard interfaces.

• The user of the application is not a highly educated expert (doctor) nor a person of general public.

• The secondary user is a client of the therapist and therefore input data to the machine learning model is provided by him/her unlike similar tools that require expensive medical equipments or sensors to obtain input data

• Cooperation with Universities., e.g. faculties of medicine, sports, mathematics, informatics.

The benefits of proposed solution

• Using of promising and popular technology can:
• add value to patients
• add value to doctors
• create database which could be used for identifying of hidden correlation between different types of disorders
• give doctors more time to deal with more demanding problems
• be in high demand
• be successful on the market.

• Using of the cloud solution enables flexibility and scalability of the application.

• Thanks to the modularity of the application Arachnet BlindPhysio can easily:
• provide more accurate results with new data
• determine diagnosis from other field of medicine
• process other type of input, i.e. video, photo or natural language
• use other type of output, i.e. natural language output

• The use of technologies that are reliable enough can:
• improving the level of health services and responding more quickly in the event of problems
• reduce medical care costs
• enable better access to health care

• Participation of therapist in the whole process can:
• be a way to solve legal problems connected to using tools based on artificial intelligence as therapist should be trained appropriately
• be a correction to underestimation or overestimation of the problems by the client/patient
• increase general confidence in artificial intelligence and modern technologies

• The cooperation with Universities can:
• help find solutions to problems that rely on highly specialized theoretical knowledge
• propose alternative and more progressive methods
• detect aminimise proposed solution errors
• help to get access to trusted medical data needed for training of Machine Learning Model of Arachnet BlindPhysio ML Model

Our visions to the future

• Processing patients' data gradually increasing in the database using the data mining techniques.

• The Arachnet BlindPhysio input can be enriched with additional elements:
• photos of patient's posture that can be specific for the specific musculoskeletal disorders
• videos of patient's way of walk that can be also specific for specific disorder

• Replacement of the currently proposed input:
• input based on Natural Language Processing, the application would process conversation between therapist and patient using added Machine Learning Model

The benefits of proposed visions

• Processing patients' data can:
• identify patterns in datasets to assess risks
• help predict falls in elderly patients
• analyse data to provide information such as severity, location and date of a disease outbreak

• Adding additional input types like photo or video of the patient can:
• give more accurate results in determining diagnosis
• give a lot of possibilities in valuable predictions, e.g. predicting of perspectives of professional athletes and their tendencies to injuries or disorders

• Using the NLP type of input can:
• be a high valued accessory to the system like telemedicine or fully automated systems of medicine

Project team

• Jan Mura is visually impaired and has knowledge and experience in the field of:
• development of large systems for telecommunication companies in Czech republic and Germany
• development of large system for insurance company
• communication with customers from many business fields
• IT technologies, i.e. RDBMS, IP, AI wide used technologies like PyTorch or TensorFlow based on Python
• analysis of customer requirements, see here
• SW engineering, see here

• Adrian Corba has knowledge and experience in:
• financial management of companies from many businesses fields
• project management in different projects in the Slovak republic and abroad
• operational management
• EU funds - sourcing and management
• Daily operational management
• Processes coordination

• Jitka Koropecka has knowledge and experience in:
• web development and design
• managing an e-shop with medical devices
• managing massage studio with visually impaired physical therapists
• dealing with legal issues of website operation, commercial law and consumer law

Our investments

• What are we investing in Arachnet Project:
• Our abilities and experience
• our enthusiasm
• Our time (we have invested a lot of it by this point)
• fully set Oracle Cloud

• For some of us, i.e. Jan Mura Arachnet is the only project he takes part in and he devotes all his time to it.

• Adrian Corba takes part also in the other projects.

Other stakeholders

• The partners involved in Arachnet:
• We are in contact with Oracle Company and we have asked them to assess the project and we hope that Oracle will support us in our efforts

Project status

• The idea of the Arachnet Project has been subjected to a thorough examination by the creators of the project.

• The AI/IT concept of the project was approved by experts from Faculty of Electrical Engeneering of Czech Technical University in Prague.

• The medical concept of the project was approved by experts from Faculty of Medicine of Comenius University in Bratislava.

• The project is ready to development.

• User requirements are defined.

• Detailed design is in progress.

• Development environment is ready.

• The tasks status can be seen here.

Costs and needs

• Time estimation of the certain parts of the project can be found here.

Conclusion

• We believe that:
• our project is innovative
• our project is inventive
• our project can arouse interest of people in many ways