Chatbot For Disease Prediction And Treatment Recommendation
Mrs. Swathi Sa, Sushruth Sb, Rajamani Rc, Manjunath Sd, Ullas Re
a Assoc.Prof, Department of Computer Science and Engineering, Nagarjuna College of Engineering and Technology,
Bangalore, India
b,c,d,e B.E. Students, Department of Computer Science and Engineering, Nagarjuna College of Engineering and Technology, Bangalore, India
Article History: Received: 10 January 2021; Revised: 12 February 2021; Accepted: 27 March 2021; Published
online: 28 April 2021
_____________________________________________________________________________________________________ Abstract: The hospital is a universal method by which any patient can undergo medical examinations, diagnose diseases and
get recommendations for any type of treatment. Most people all over the world follow this custom. People have considered this to be the most suitable and authentic way to keep a check on their physical body state. The proposed system is to find another alternative to this formal way of having to visit a hospital and to make an appointment for a check-up or diagnosis with a doctor. This particular research study will help to apply the concepts of natural language processing and machine learning to create a chatbot application. Just like people interact with each other, here people can easily interact with the chatbot through a series of questions and doubts, meanwhile the chatbot will find and identify the person's symptoms and thereby can predict which disease the person is affected with and suggest the related remedies and treatment. This system can be proved to be of great use especially to people who have to conduct diurnal check- ups, It can also enable people to understand their health and encourage more people to take appropriate measures to maintain their health. This research also suggests that such a system is not very popularly used and people lack knowledge about this. Putting into action this framework can be of great help to people in avoiding long-distance trips to the hospital just by using this free app no matter where you are.
Keywords: medical scrutiny, disease prognosis, chatbot, diurnal check-ups
1. Introduction
A flourishing society is a point at which the entirety of its individuals is solid. It is imperative to keep up wellbeing if one wishes to be glad. Just a solid body can have a sound mind and positively affect individuals' presentation. In their bustling lives, they neglect to take fitting measures to keep themselves sound and are less mindful of their wellbeing status. In the most recent news from TOI [1], we can see that individuals don't join significance to their wellbeing that testing in clinics takes too long. A bustling planned life has a bad situation for well-being. The majority of individuals who make up the local area's branch of work claim that their furious timetable doesn't allow them to have normal clinical checks and that they overlook any nervousness their bodies show until it turns out to be excessively extreme.
The principal motivation behind the task is to make the language hole between the client and wellbeing specialist organiza- tions by giving prompt reactions to the in- quiries posed by the client. There are three analyzes of tongue comprehension, that is, the completeness of identifying the main lin- guistic relationships for analyzing the topic in the subject of sentences. At that point, the portrayal of the writings is finished. Seman- tic understanding uses knowledge of the word that implies Chatbot is a substance that mirrors hu n banter in its satisfactory setting alongside a book or phonetic language (NLP). The objective of this framework is to
rehash the individual's conversation. Regu- larly the chatbot application occasion winds up making an interface for sending inputs and accepting a reaction. It is a framework that interfaces with the client by following the situation with the connection and recol- lecting past orders to give usefulness. Clini- cal chatbots are regularly evolved utilizing manufactured calculations that examine cli- ent requests, remember them, and give the response to the pertinent query framework gives response utilizing a proficient graph- ical interface like if the genuine individual is conversing with the client. A chatbot can be utilized in different fields like instruction, medical care, and emergency aides. The Central District of Chatbots incorporates MySQL. It is an intuitive framework that set- tles clients' requests identified with medica- tion. With the goal that they get the right guidelines for treatment through the web ap- plication utilizing Google API.
2. Literature Survey
In the paper by Rashmi Dharwadkar [5], the working of a chatbot relies upon Natural language preparation that causes clients to advance their issues about actual wellbeing. The patient can ask his wellbeing- related issues/inquiries through the clinical chatbot, it isn't required to test the client/patient should mandatorily go to the medical clinic rather by utilizing Google API for text-voice or voice-text discussion. Chatbot gets the inquiry from the client and showcases the connected arrangement through an android app.
The proposed idea of the paper "A novel approach for medical assistance using trained chatbot" by Divya Madhu [4] is to plan a model utilizing computerized reasoning eat encourages the client to perceive the certified treatment for sickness. There are a ton of medicines accessible for a specific illness and nobody can explicitly recommend the appropriate treatment and which is the best counterpart for that infection. In this proposed, man-made brainpower assumes a significant part by giving a record of accessible medicines
dependent on the illness perceived through the side effects. The framework recommends the medications for illnesses and their rec repaired utilizes and causes the client to choose an appropriate treatment. This framework urges individuals to Ave a fundamental thought of their wellbeing and monitor their health status, accordingly encourages the client to take appropriate treatment.
The proposed thought of the paper "A self-conclusion clinical chatbot utilizing man-made reasoning" by S. Divya [6] is to develop a framework utilizing man-made consciousness subsequently causes clients to approach the doctor from far off zone where he leaves, and the client need not counsel the doctor. This framework is made to perceive the sickness of the client and think of required insights regarding the infection. This framework is more practical including enhancing terminate to information about the sickness. This chatbot is legatee to clients just when it can distinguish the kind of illness and give important subtleties to the client. The proposed system is an agreeable one that reacts with clients to recuperate knowledge about the ailments, as such giving a legitimate finding.
In the paper by Amiya Kumar Tripathy [7], it indicates that progression in innovation is required, in this manner by giving appropriate administration of the care framework, where individuals become accustomed to it rather than a specialist. It focuses on the need for such sort of frameworks to be more exact and it ought to be versatile so clients can convey it starting with one spot then onto the next. The proposed framework involves a portable pulse estimation where a pulse can be recognized and dependent on this, legitimate treatment will be furnished with a tick of a catch. This framework gives innovation sucks video conferencing so a patient can associate effortlessly with a crisis. The Doc- Bot that was created for the design being utilized is currently changed over a versatile
stage while it has consolidated a thought of giving prescriptions dependent on the manifestations.
In the paper by S. du Preez [8], it put advances the possibility of an insightful voice acknowledgment chatbot. The improvement of the framework which is used is proposed by the plan of the framework and the essential innovation identified with the turn of events. It gives a web-administration that allows any customer to get to the worker from anyplace. It utilizes the idea called the discovery way to deal with control the correspondence, back and forth from the web-administration. This is reachable through an interface that permits consistent xml handling. By the utilization of a counterfeit mind, the electronic chatbot reacts to the client inquiries, if the client doesn't see any journey s, the following stage will be advanced utilizing an online wise examination collaborator and the reactions will be recorded to give methodical reactions later on.
The paper by B. R. Ranoliya {9], characterizes chatbots as projects that can copy human conversation using artificial knowledge. E proposes the possibility of a chatbot as a menial helper or as a keen specialist that can tackle assignments like giving legitimate reactions to questions from clients, controlling gadgets, giving courses during driving, and so on The paper presents the possibility of a chatbot, Each helps in giving a sensible reaction to the inquiries that were asked by the client, in light of the dataset of FAQs. For this reason, it consolidates both man-made consciousness increase language and inert septic examination. These kinds of dialects are utilized to react to a particular type of question. Artificial insight increases language is utilized to recognize format- based and general inquiries. Reactions to other assistance-based inquiries are given utilizing dormant semantic investigation.
This kind of language is predominantly utilized in colleges to cooperate with understudies while answer FAQs to them.
Fig 1: Methodology
The system architecture is a mental model that explains the system's structure and actions. It involves the system components as well as the relationships that explain how they interact to execute the overall system. In the block diagram In the above fig the user ask questions in the chatbot as the front end consists of the user interface of the user input requests of queries of the patient and the chatbot, request queries are sent back end in which the chatbot sever in bot controller logic it controls what happens with a message automatically to the user has to the requested queries from the chatbot client and according to it checks data from the pre-processing model where the raw data is taken and prepared according to the user model and gives accurate responses to the chatbot client. In SQLite, the dataset is a single row of data, instance, and collection data whichever we have trained to the datasets according to request in the chatbot server where trained data and test data is stored in the SQLite sever which feed based on the details of the user. In case of the absence of key patterns, the virtual doctor prescribes the medicines based on the symptoms and using machine Medical chatbot learning logic that is support vector machine (SVM) algorithm which identifies the disease based on the symptoms given by the user in the chatbot and predicts the diseases and analyses the disease and prescribes the medicine based
on the age of the particular patient. The objective is to predict the diagnosis of disease with several attributes and provides a solution to the patient through a chatbot. This project aims to use a classifier model to recognize key trends or features in medical data. Classification is used to predict the diagnosis of disease after the reduction of the number of attributes of the user.
Fig 2: Usecase diagram of chatbot
Fig 3: Architecture of chatbot
Fig 4: Sequence Diagram 4. Software And Languages Used
each other, and programming could be done. So, this is considered a huge lead for jupyter. Data scrubbing and conversion, computational analysis, analytical designing, predictive analysis and many further can be applied in this. To run the algorithmic program IPython is used. The inter mutual estimating domain called IPython Notebook allows clients to examine logbook(notebook) files that comprise: Programming, Shared gadgets, Graphs, Chronicle wordings, Calculations, Pictures, Visual communication pictures. To transform for several formats along with we need to interact with the rest of them using Email, Dropbox, these notebook files provides an outright and independently preserved data of an enumeration.
The IPython Notebook constitutes of trio elements namely, The logbook(notebook) network application, Kernels(Programming languages), Notebook files. The notebook network application is a shared network application meant for scripting and running code mutually one another and penning notebook files.
Kernels(Programming languages) are the disparate procedure initiated by the logbook network application that executes the code in a given dialect and gives back output away to the notebook network application. Computing for inter mutual gadgets, tab completion, and self-analysis are the objects which can hold by the kernel. Notebook files: Collected files that hold a portrayal of all information enable in the notebook network application, constitutes of inputs and outputs
of the enumerations, chronicle wordings, calculations, pictures, and multimedia portrayal of things. Each notebook file has its kernel.
Python
Python is an interpreted high-level programming language. The creator of this language is Guido van Rossum and it was initially launched in 1991. Python has a pattern ideology that highlights programming legibility, particularly using notable blank space. This is the familiar coding language in current days. It gives a set up to authorize understandable coding on both tiny and wide range extensions. The logical retreat in the method is executed in jupyter and the python language is used for a writeup of the algorithm.
PyCharm
PyCharm is an IDE used in coding, particularly for the Python language. Czech company JetBrains is a web emergent in this. Static analysis, plots, or graphs troubleshooter, consolidated code probing will be provided by the PyCharm. Interface design with Django in addition to data science with Anaconda will also be supported by PyCharm.
Developers can create their add-in to enlarge PyCharm characteristics if PyCharm provides an API. Various add-ins from other company integrated development environments to function with PyCharm. There are beyond thousand add-ins that are well-matched with PyCharm.
Natural language processing[NLP]
Natural language processing [NLP] is a dimension of the study of language, computer-aided learning, and expert systems affiliated with the interconnection between mainframe and person dialect, specifically how to arrange computers to do an activity and inspect abundant written language data. The solution is a computer competent of
"understanding" the information of files, in addition to the contingent shade of the dialect within them. The innovation later precisely removes data and the ability to accommodate in the files along with classifying and systematize the files themselves.
NLP converts person dialect into data with a union of design and words that can be outlined immediately to spot relevant solutions. There are NLP contributions and APIs that are used to assemble the chatbots and can be implemented for all kinds of business venture works and small things as well.
Naïve bayes algorithm
It is a categorizing approach that depends on Bayes' rule with an acceptance of liberty among analysts. A Naive Bayes classifier thinks that the existence of a specific attribute in a group is irrelevant to the existence of any other characteristic.
The trained data is not required for this algorithm. It handles both uninterrupted and disconnected data. It is extremely ascendible with the lots of analysts and data points. It is rapid and it is well used to make real-time predictive analytics.
Fig 5: Flowchart of NB algorithm 5. Conclusion And Future Scope
The future scope of the system is to develop a different method for this traditional type of hospital visits and appointments for doctor consultations to get diagnosis by designing and developing a tool like a medical chatbot for health using machine learning algorithms and NLP (Natural Processing language). The objective of the system is
• To provide chatbot service for pregnant ladies by using Natural Language Processing (NLP). To recognize the key patterns or aspects from the medical data with the help of the classifier model.
• To predict, analyze and provide a solution to the patient using different attributes through the chatbot.
References
1. https://timesofindia.indiatimes.com/life style/health-fitness from “The Times of India”
2. “Survey on Chatbot Design Techniques in Speech Conversation Systems” by Sameera A. Abdul-Kader, Dr. John Woods from School of Computer Science and Electronic Engineering/University of Essex Colchester/ UK.
3. “Commercial Chatbot: Performance Evaluation, Usability Metrics and Quality Standards of Embodied Conversational Agents” by Kuligowska, Karolina (2015). Professionals Canter for Business Research. 2. 1-16. 10.18483/PCBR.22.
4. Divya Madhu, Neeraj Jain C. J, Elmy Sebastain, Shinoy Shaji, Anandhu Ajayakumar. A Novel Approach for Medical Assistance Using Trained Chatbot, International Conference on
5. Inventive Communication and Computational Technologies (ICICCT 2017).
6. Mrs. Rashmi Dharwadkar, Dr.Mrs. Neeta A. Deshpande "A Medical ChatBot" in International Journal of Computer Trends and Technology (IJCTT) V60(1):41-45, June 2018.
7. S. Divya, V. Indumathi, S. Ishwarya, M. Priyasankari, S. Kalpana Devi, "A Self- Diagnosis Medical Chatbot Using Artificial Intelligence", J. Web Dev. Web Des., vol. 3, no. 1, pp. 1-7, 2018.
8. Amiya Kumar Tripathy, Rebeck Carvalho, Keshav Pawaskar, Suraj Yadav, “Mobile based healthcare management using artificial intelligence” in International Conference on Technologies for Sustainable Development (ICTSD), 2015, 4-6 Feb, 2015.
9. S. du Preez, M. Lall, S. Sinha, "An intelligent web-based voice chat bot", EUROCON 2009 EUROCON'09. IEEE, pp. 386-391, 2009.
