Security researchers at Ben-Gurion University Cyber Security Research have created new software designed to determine whether an attacker could hack CT and MRI scanning equipment to produce false results for tumor patients. The findings were positive, as the tampering was not difficult to achieve.
According to a Washington Post on the subject, “ attackers could target a candidate or other politicians, tricking them into believing they have a serious illness and forcing them to withdraw from the campaign to seek treatment.”
Radiologists can be tricked by the malware into seeing fake cancer cells when none are present.
The malware was actually created by researchers in Israel. They wanted to explore security vulnerabilities in both medical imaging equipment and the networks that transmit those images.
The researchers' malware could act in two ways – either by adding fake tumors to the scans or by removing real tumors. The latter could obviously lead to failure to cure patients in need of immediate treatment.
The malware altered 70 images and managed to trick three radiologists into believing the patients had cancer. The researchers used lung cancer for the most part.
In cases where the malware hid actual cancer cells from scans, radiologists said the patients were healthy 94 percent of the time.
The study's results should prompt the medical community to consider this issue in terms of potential impact: The attackers' motives could be general or targeted. They could simply want to cause chaos, or they could use the malware to target specific patients.
Malware could also have adverse effects on drug trials and medical research “to sabotage the results.”.
BBC News reported that hospitals and other healthcare organizations have been a popular target for cyber attackers, and many have been hit by malicious ransomware (such as WannaCry) that encrypts files and returns the data only when victims pay.
Why is malware able to bypass any security gateways? The Washington Post reported that the problem could be traced to the equipment and networks that transmit and store CT and MRI images.
“These images are sent to radiology workstations and back-end databases via a picture archiving and communication system (PACS). The attack works because hospitals do not use encryption on their PACS networks, thus allowing an attacker on the network to view the scans and alter them.”.
PACS networks are generally not encrypted. Another potential problem mentioned in the article lies in hospitals dealing with “20-year-old infrastructure” that does not support newer technologies.
"In this paper we presented the possibility of an attacker modifying a 3-D medical image, using deep learning. We presented a manipulation framework (CT-GAN) that can be executed by a malware autonomously.".
Additionally, “The altered images were also able to fool automated viewing systems,” BBC News reported.
