FIELD OF THE INVENTION
The present invention herein relates to the field of chemistry information technology particularly to a method of virtual medical prescription based on machine learning and artificial intelligence.
BACKGROUND OF THE INVENTION
There is a growing shortcoming of physicians in rural areas leading to unattended patients. One solution to overcome this problem in future is the use of AI as a healthcare system. This AI will determine the diseases the patient is suffering from and in latter use can provide appropriate medicine or treatment details to the patient. Using automated system of AI, the patient will be able to get quick and accurate details thus saving time and money.
Following are some most relevant prior arts:
US20130024209A1; A user can get up-to-date medical information directly from an interactive network-based health information system. The data is suited to the user's level of competence. The user can ask precise follow-up questions, initiate a professional conversation, and form a doctor-patient relationship. The device allows for patient monitoring and diagnosis as well as therapy from a distance. The various degrees of service can be given and charged individually.

CN105144171B
In certain ways, a method of assisting medical professionals utilising Virtual Medical Assistant, in which the Virtual Medical Assistant is at least partially realised by at least one processor and is connectable to the host equipment of at least one network. Receiving free form instruction from a medical practitioner is part of the process. The medical professional described in the forward direction for executing at least one medical task provides at least one first response instructed about the free form and receives the first information from the medical professional in response to at least one described first response.
None of the prior art indicate above either alone or in combination with one another disclose what the present invention has disclosed.
SUMMARY OF THE INVENTION
This summary is provided to introduce a selection of concepts, in a simplified format, that are further described in the detailed description of the invention.
This summary is neither intended to identify key or essential inventive concepts of the invention and nor is it intended for determining the scope of the invention.
This method of the present invention provides the support for both doctor as well as patient. It will provide the measures to treat the illness based on the symptoms given by the patient either by text or voice input. It will also provide the count of patient with their details in advance for OPD purpose to doctor also. So that doctor can schedule all meetings in advance.
The present invention provides a method of virtual medical prescription based on Machine learning and Artificial Intelligence that does not require the human intervention for detection of an illness. The present invention will also provide the measures to treat the illness based on the symptoms given by the patient either by text or voice input. A machine learning module is trained to accurately determine to whether an outcome in the computing system resulting from the input, satisfy an outcome threshold and collusion. It involves the use of training data set in ML for general or common illnesses. A neural network architecture based on the patient new health data and previous data to provide a modified neural network architecture is used. The machine learning system improves its accuracy by continuous training of a backend machine learning model based on explicit tag signals gathered from monitoring of user interaction during a review process. In this model the AI will be able to interact with the patient better with the help of speech recognition and animation and thus will give user the experience of real person.
To further clarify advantages and features of the present invention, a more particular description of the invention is rendering by reference to specific embodiments thereof, which is illustrated in the appended drawing.
It is appreciated that the drawing depicts only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention is being described and explained with additional specificity and detail with the accompanying drawing.

BRIEF DESCRIPTION OF DRAWINGS
The foregoing detailed description of embodiments is better understood when read in conjunction with the attached drawing. For better understanding, each component is represented by a specific number which is further illustrated as a reference number for the components used with the figures.
With the help of figures the detailed explanation of connections become easier and understandable. For easier understanding each component is represented by a reference numeral which is also used as a reference number for the components used in the illustration.
Figure 1 represents deep neural network
Figure 2 represents the block diagram of the virtual system for medical prescription
Further, skilled artisans will appreciate that elements in the drawings are illustrated for simplicity and may not have necessarily been drawn to scale. Furthermore, in terms of the construction of the system, one or more components of the system may have been represented in the drawing by conventional symbols, and the drawing may show only those specific details that are pertinent to understanding the embodiments of the present invention so as not to obscure the drawing with details that will be readily apparent to those of ordinary skill in the art having benefit of the description herein.

DETAILED DESCRIPTION OF THE INVENTION
For the purpose of promoting an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawing and specific language will be used to describe the same.
It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skilled in the art to which this invention belongs.
Embodiments of the present invention will be described below in detail with reference to the accompanying drawings.
The availability of primary medical care is strongly limited in rural areas due to low population densities. This problem requires long distance trips to take necessary precautions or treatment from medical professionals and the ineffective allocation of time and resources makes it difficult for medical professionals to provide services to the patients of Urban areas. Most larger cities have low general practitioners to population ratios due to population density. This results in a high number of patients, which have to be seen by each practitioner on daily basis, lowering time and resources available for each patient. These problems of current medical care could all be improved by a partially autonomous AI for anamnesis and for medical diagnosis. In highly frequented general practitioners, the time-consuming process of anamnesis could be automated and thus standardized. The AI could serve as a start point of care for patients with enhanced medical assistance. The AI could provide the patient to the closest service according to the situation or advise to see a practitioner in some critical situation. By the automation of the basic elements required in types of medical care, anamnesis, initial assessment of situation and diagnosis, it would be possible to provide such a service in a very standardized fashion available at any time.
This model uses ML/AI to be used in healthcare application to increase or boost clinical activity. Machine learning involves the study of computer algorithms that can improve automatically through experience or use of data. It has the capability to imitate intelligent human behaviour and work differently according to the situations. ML is used to easily identify trends and pattern with no or less human intervention. This method uses reinforced learning that involves an algorithm that improves itself and learns from new situations using a trial-and-error method. Based on the concept of conditioning, reinforcement learning work by an algorithm in an environment with an interpreter and a system. The reinforced learning is integrated with adaptive algorithm and sub-sampling approach to diversely see the illness through symptoms. The decision system is used to give the actual disease and its causes. It involves the use of deep neural network which helps in determining the score of effectiveness based on probability estimation code. The system works on a score of effectiveness, expressed in a percentage value. The higher the percentage the more favorable the outcome will be. Then after calculation of results it will provide a report of the illness the patient is suffering and tell to consult physician in case an immediate treatment is required.
Methodology:
The aim is to create an interactive virtual system for medical prescription that will use AI integrated with other co-related concepts to examine the patient health as well as suggest its treatment according to its system altogether with connected neural networks and previous data set. The model will take the symptoms of the patient illness either by text or speech recognition and then accordingly will suggest the treatment to the patient.
Detection of problem: An interactive computer assisted method review and analyzes symptoms associated with the patient. Symptoms will be given as a user input with the help of text or speech recognition module to be submitted. In order to detect the problem, it will go through neural network and give the best suited problem as well as treatment for the problem.
Speech Recognition: We will use speech recognition as an additional component to interact with the patient in order to facilitate the process. In latter use, speech recognition can be improved in order to fit for different languages that will provide the user with better experience.
Machine Learning: Machine learning is used to provide the module the experience to accurately detect or informed about the illness. ML is used to find pattern in the new input set and then give output based on prediction. The machine learning involves certain steps to follow:
• Gathering and preparing data
• Choosing and training the model
• Evaluation
• Hyperparameter Tuning
• Prediction Process
Algorithm: Algorithm used is a type of adaptive algorithm that is a continuous algorithm which changes its behavior as result of learning from previous input over time. By repeatedly learning and updating, adaptive algorithms will have a vast potential in terms of optimizing the performance and accuracy of medical tech. while processing. Adaptive algorithm is capable of machine learning that are used in medical devices saves time, reduced errors and improved user/patient care. Another approach is integrated into the previous algorithm which is based on sub-sampling approach. By using the sub-sampling approach, a number of times, the system will be able to increase the unbiased-ness in the dataset and thus able to view the illness more diversely.
Decision Support System: The use of decisions support system provides a significant paradigm shift characterized by a cycle of systematic links and mutual feedback between bigdata acquisition and modelling. The system will also interact with the physician in order to analyze the patient data until it reaches a certain level of accuracy to get additional conditions and exceptional cases into the dataset. In our model we will use deep neural network for the prediction of the problem. Deep learning will provide the model a human like intelligence that will increase the efficiency of the model and will be trained by the previous set of data. So, the model will be more accurate in any condition that may come ahead in the processes. We will test the neural network from 1 to 10 neurons in the hidden layers.
Probability Estimation: The probability estimation of estimation of illness will be dependent on system on the well calibrated prediction. A well calibrated prediction can be defined as an estimation if an event actually happens with an observed relative frequency consistent with the previous prediction. The system works on a score of effectiveness, expressed in a percentage value. The higher the percentage the more favorable the outcome will be. This also uses a probability density filling algorithm based on highest log-likelihood and returns calibrated predictions.
Animation: The animation of the doctor will be provided to make it more real looking as to provide better user experience. In the latter use, hologram of the doctor can also be used if paired with other required devices.

We Claims:

1. A method of virtual medical prescription based on machine learning and artificial intelligence; comprising the steps of;
detection of illness by analysing symptoms associated with the patient by input to computer module with the help of text or speech recognition module;
wherein, Machine learning is used to accurately detect about the illness and to find pattern in the new input set and then give output based on prediction;
a decisions support module provides a significant paradigm shift characterized by a cycle of systematic links and mutual feedback between bigdata acquisition and modelling;
estimation of probability of illness.
2. The method as claimed in claim 1, wherein, detection of illness based on symptom analysis, will go through neural network and give the best prediction of illness as well as treatment for the illness.
3. The method as claimed in claim 1, wherein, the decision support module interacts with the physician in order to analyze the patient data until it reaches a certain level of accuracy to get additional conditions and exceptional cases into the dataset by using deep neural network for the prediction of the illness.