DEEP NEURAL NETWORK-BASED INTRUSION DETECTION SYSTEM FOR MEDICINE VENDING MACHINE

Abstract

The present invention relates to a deep neural network-based intrusion detection system designed for securing an automated medicine vending machine. The system comprises a network traffic monitor that captures and preprocesses network data, a deep neural network (DNN) module trained to identify anomalous patterns indicative of cyberattacks, and a user authentication interface utilizing biometric or password-based access for secure user verification. An automatic medicine allotment module is configured to dispense prescribed medication upon successful user authentication, while a machine learning interface continuously updates the system's detection parameters to adapt to emerging threats. The invention also includes an alert and response system to prevent unauthorized access and ensure safe medication distribution. The proposed solution provides a secure, automated method for dispensing medication, mitigating risks associated with unauthorized access or cyberattacks.

Specification

Description:[013] The following sections of this article will provide various embodiments of the current invention with references to the accompanying drawings, whereby the reference numbers utilised in the picture correspond to like elements throughout the description. However, this invention is not limited to the embodiment described here and may be embodied in several other ways. Instead, the embodiment is included to ensure that this disclosure is extensive and complete and that individuals of ordinary skill in the art are properly informed of the extent of the invention. Numerical values and ranges are given for many parts of the implementations discussed in the following thorough discussion. These numbers and ranges are merely to be used as examples and are not meant to restrict the claims' applicability. A variety of materials are also recognised as fitting for certain aspects of the implementations. These materials should only be used as examples and are not meant to restrict the application of the innovation.
[014] Referring now to the drawings, these are illustrated in FIG. 1, the proposed deep neural network-based intrusion detection system is integrated into a medicine vending machine to detect and prevent unauthorized access attempts. The system comprises the following main components:
Network Traffic Monitor
This module captures all incoming and outgoing data associated with the medicine vending machine. The data includes user authentication credentials, medicine inventory records, and prescription information. The network traffic monitor acts as the primary point for data acquisition, allowing the system to have a comprehensive overview of all interactions with the vending machine. It filters and preprocesses the data before forwarding it to the deep neural network for analysis. The data filtering process removes noise and irrelevant information, while preprocessing involves normalizing the data and transforming it into a suitable format for analysis. The module also ensures that any suspicious activities, such as repeated failed login attempts, unusual access times, or abnormal data patterns, are flagged for further inspection and forwarded to the intrusion detection module.
[015] In accordance with another embodiment of the present invention, the DNN module processes the network traffic in real time to identify anomalous patterns indicative of a cyberattack. The neural network has multiple layers, including input, hidden, and output layers, which are trained using a large dataset containing both normal and malicious network behavior patterns. The input layer receives the preprocessed data from the network traffic monitor, while the hidden layers extract complex features to differentiate between normal and malicious activities. The hidden layers utilize activation functions such as ReLU (Rectified Linear Unit) to introduce non-linearity, making the system capable of capturing complex attack patterns. The output layer uses a softmax function to classify the activity as either legitimate or a potential threat. The DNN continuously learns from new data using techniques such as gradient descent and backpropagation, making it highly adaptive to new and sophisticated attack patterns. The training dataset is periodically updated to include new attack vectors, ensuring that the DNN remains effective in detecting emerging cyber threats.
[016] In accordance with another embodiment of the present invention, the vending machine is equipped with a user authentication system that utilizes biometric data, such as fingerprints or facial recognition, or password-based access to verify the identity of the user. The biometric sensors are connected to the system through secure communication protocols to prevent tampering. Upon successful authentication, the machine allows the user to proceed with selecting and obtaining prescribed medicines. The user authentication interface is directly linked to the intrusion detection system, ensuring that only verified users can interact with the machine. If the DNN detects any anomalies during the authentication process, such as an unauthorized user attempting to bypass the authentication by spoofing biometrics, it immediately halts the process and triggers an alert. The user interface also logs each authentication attempt, providing a detailed record of access attempts, which is analyzed for any unusual patterns.
[017] In accordance with another embodiment of the present invention, once the user is authenticated, the vending machine dispenses the prescribed medication. The system cross-references the user's prescription with available inventory and dispenses only the authorized dosage. This module interacts with both the user authentication interface and the intrusion detection module to ensure that no medication is dispensed if any suspicious activity is detected. The allotment module uses a stepper motor-based dispensing mechanism that ensures precise control over the amount of medication dispensed. The allotment process is monitored by sensors to detect any physical tampering or attempts to access the dispensing mechanism without proper authentication. The allotment module also maintains a log of all transactions, which is monitored by the DNN to detect any irregularities, such as unusually high requests for medication that could indicate misuse or an attempt to exploit the system as shown in figure 2.
[018] In accordance with another embodiment of the present invention, the machine learning module is responsible for continually updating the intrusion detection parameters. By analyzing previous incidents, the module adjusts the DNN's detection capabilities, thus adapting to new and sophisticated threats over time. This interface collects feedback from past detections and false alarms to fine-tune the model, thereby improving its overall accuracy. The machine learning interface uses reinforcement learning techniques to adjust the weights and biases of the DNN, ensuring that it learns from both successful and unsuccessful intrusion detection attempts. The machine learning interface also works closely with the network traffic monitor to identify new patterns of network behavior, ensuring that the system remains up-to-date with evolving cyber threats. Data from network traffic is fed back into the model, enabling unsupervised learning for anomaly detection and clustering of new threat patterns.
[019] If an intrusion attempt is detected, the system triggers an alert. The alert can be communicated to the network administrator through a secure channel, such as an encrypted email or SMS, while the machine itself will halt the dispensing of medication until the threat is resolved. The alert and response system is designed to take immediate action to prevent unauthorized access. This includes locking the vending machine using an electromechanical lock, notifying relevant authorities, and logging the details of the detected intrusion, including the time, nature of the anomaly, and the actions taken by the system. The response system also has a self-check mechanism that verifies the integrity of the system components after an intrusion attempt, ensuring that the machine is safe to resume normal operations. The self-check mechanism uses a series of diagnostic tests to verify that no components have been tampered with, and it performs a checksum validation to detect any unauthorized modifications to the software.
[020] The proposed invention offers multiple advantages, including:
Enhanced security through real-time intrusion detection based on deep learning algorithms.
Increased adaptability to emerging threats using machine learning interfaces.
Secure and efficient dispensing of medicines only to authenticated users.
Reduction in false positives and false negatives for intrusion detection, ensuring a reliable and safe user experience.
[021] This invention can be applied to healthcare facilities, pharmacies, and public health systems to provide secure and reliable access to medications through automated vending machines, while ensuring that cybersecurity threats are proactively addressed. The system can also be utilized in remote or rural areas to provide medication access with integrated security features, enhancing healthcare outreach.
[022] The proposed deep neural network-based intrusion detection system provides a robust solution for securing medicine vending machines against cyberattacks, thereby ensuring the safe and reliable allotment of medications. By leveraging advanced machine learning techniques and biometric authentication, the system enhances both security and usability for healthcare providers and patients alike.
[023] The benefits and advantages that the present invention may offer have been discussed above with reference to particular embodiments. These benefits and advantages are not to be interpreted as critical, necessary, or essential features of any or all of the embodiments, nor are they to be read as any elements or constraints that might contribute to their occurring or becoming more evident.
[024] Although specific embodiments have been used to describe the current invention, it should be recognized that these embodiments are merely illustrative and that the invention is not limited to them. The aforementioned embodiments are open to numerous alterations, additions, and improvements. These adaptations, changes, additions, and enhancements are considered to be within the purview of the invention. , Claims:1. A deep neural network-based intrusion detection system for a medicine vending machine, comprising:
a network traffic monitor configured to capture and preprocess incoming and outgoing data associated with the vending machine;
a deep neural network (DNN) module trained to identify anomalous patterns indicative of a cyberattack;
a user authentication interface that utilizes biometric data or password-based access for user verification;
an automatic medicine allotment module configured to dispense prescribed medication upon successful user authentication and includes sensors to detect physical tampering or unauthorized access to the dispensing mechanism and
a machine learning interface configured to update the intrusion detection parameters continuously;
wherein the network traffic monitor filters and preprocesses data, removes noise, and transforms the data into a suitable format for analysis;
wherein the deep neural network module comprises multiple layers, including input, hidden, and output layers, configured to classify activity as either legitimate or a potential threat and uses a training dataset containing both normal and malicious network behavior patterns to differentiate between normal and malicious activities.
wherein the deep neural network module is configured to continuously learn from new data and update its detection capabilities using techniques such as gradient descent and backpropagation.
wherein the user authentication interface comprises biometric sensors configured to authenticate the user via fingerprint or facial recognition, and is connected to the system through secure communication protocols;
wherein the automatic medicine allotment module uses a stepper motor-based dispensing mechanism that ensures precise control over the amount of medication dispensed.
2. The system as claimed in claim 1, wherein the automatic medicine allotment module is configured to cross-reference the user's prescription with the available inventory and dispense only the authorized dosage.

3. The system as claimed in claim 1, wherein the machine learning interface utilizes reinforcement learning techniques to adjust the weights and biases of the deep neural network, thereby improving detection accuracy over time.

4. The system as claimed in claim 1, further comprising an alert and response system configured to trigger an alert upon detection of an intrusion attempt, halt medication dispensing, and communicate the alert to a network administrator through a secure channel.

5. The system as claimed in claim 1, wherein the alert and response system includes an electromechanical lock to secure the vending machine upon detection of an unauthorized access attempt.

6. The system as claimed in claim 1, wherein the alert and response system is configured to log the details of the detected intrusion, including the time, nature of the anomaly, and actions taken by the system.

7. The system as claimed in claim 1, further comprising a self-check mechanism that performs diagnostic tests and checksum validation to verify the integrity of system components after an intrusion attempt.

8. A method for securing an automated medicine vending machine using a deep neural network-based intrusion detection system, comprising the steps of:
a) capturing and preprocessing network traffic data associated with the vending machine;
b) analyzing the preprocessed data using a deep neural network to identify anomalous patterns indicative of a cyberattack;
c) authenticating the user through a biometric or password-based user authentication interface;
d) dispensing medication based on user authentication and prescription verification; and
e) updating the intrusion detection parameters continuously using a machine learning interface to adapt to new threats.

9. The method as claimed in claim 8, further comprising the step of triggering an alert and halting medication dispensing upon detection of an intrusion attempt, and notifying a network administrator through a secure channel.

Documents