<p>Due to e-health care, medical images are transmitted along with the patient details in Electronic Patient Records (EPR) over an insecure medium for further investigation and diagnosis. The EPR consists of all valuable and relevant patient information, which tempts hackers to steal and utilise it for malicious activities, resulting in healthcare data breaches and threats. Furthermore, the healthcare sector has recently been more susceptible to Denial of Service (DoS) attacks, as the Ponemon Institute report suggests. To overcome this alarming situation, a suitable intrusion detection system must be developed and deployed to detect DoS attacks, enabling healthcare management to implement effective cybersecurity. This paper proposes an intrusion detection system on Python Productivity for Zynq(PYNQ Z1)System on Chip (SoC). The lab-based customised 8,264 normal and 3,004 malicious packets were generated using the Scapy tool and captured in Wireshark. This was classified into normal and abnormal packets at the DoS detection unit on Jupyter Notebook in the PYNQ Z1 kernel, enabling access to the PACS from other authorised systems. Among the 11,268 packets generated, 7,887 (70%) were used for training and 3,381 (30%) for testing. The six different ML classification algorithms, including Logistic Regression, Support Vector Machine (SVM), K-Nearest Neighbour (K = 5), Naïve Bayes (Bernoulli), Decision Tree, and Random Forest (with 1000 Decision trees), were tested for their effectiveness. Among these, the random forest of 1000 decision trees accurately detects DoS attacks with precision, recall, and F1 scores close to unity.</p>