AI screens abnormal images, imaging cloud breaks hospital information silos, and network security architecture secures patient data in the era of big data …… Forbes has released six medical trends for the new decade, will these revolutionary technologies change the existing disease diagnosis management model? Will the future of healthcare be driven by technology to return to the original intention of “human-centered”?
AI screening of abnormal images
With the development of technology, medical imaging has become a necessary tool for doctors to make accurate diagnosis. Relevant data shows that 70% of clinical diagnosis needs to rely on medical images. Since medical images have the characteristics of being storable, transferable and relatively standardized, they have also become the earliest field to be applied in AI research and development and landing. According to the latest data from Global Market Insight, AI+medical imaging, the second largest segment of AI medical applications after AI+drug development, will grow at a rate of more than 40% to reach $2.5 billion in 2024. In addition to the cliché of assisting imaging doctors in diagnosis, how else can AI play in medical imaging?
Imaging doctors need to look at a huge number of films every day to determine whether a patient has an abnormality. If they find certain abnormalities, they mark the image as “positive” and order the patient to undergo further tests, a process that usually takes a lot of time. An Israeli medical technology company has developed a series of AI solutions that can screen for abnormal images. For example, their AI solution Aidoc, developed for stroke, can quickly and continuously identify abnormal ischemic and hemorrhagic stroke images and automatically move suspicious cases to the top of the imaging physician’s to-do list, alerting the imaging physician to diagnose them as soon as possible and speeding up the process of deciding on the best treatment option. If the AI diagnosis is correct, the imager can move on to diagnose the next case. In fact, if the AI makes a wrong diagnosis, the case becomes training data for the AI as well. Over time, the Aidoc system will become more accurate. It is understood that Aidoc’s AI solution for stroke has been approved for marketing by the U.S. FDA. In addition, the company’s AI solutions developed for pulmonary embolism, cervical fracture and intracranial hemorrhage have all been approved for marketing by the FDA and have been put into clinical use in more than 300 healthcare facilities worldwide.
Cloud computing to break the hospital information silo
According to a recent report by International Data Corp, IT infrastructure spending for cloud deployments is expected to exceed $40.1 billion this year. $40.1 billion of this huge spending accurately reflects the confidence in cloud computing as the computing model of the future and the trend of cloud computing models catching fire. For healthcare, cloud computing is not just about having large amounts of computing resources at a lower cost; it is about adopting an approach that will fundamentally improve the way healthcare organizations communicate, collaborate and conduct business.
Most healthcare organizations worldwide currently use the DICOM (Digital Imaging and Communications in Medicine) standard for their medical imaging systems (PACS). However, due to the closed and heterogeneous nature of hospital systems, the problem of medical imaging data compartmentalization and silos is still serious, which not only hinders the sharing of medical imaging data and the establishment of personal medical imaging electronic health records, but also adds difficulties to the construction of medical imaging-based intelligent medical services. The establishment of a regional cloud for medical imaging can obviously solve the problems of multi-territory storage of imaging data and cross-regional data backup, and realize regional image sharing. The sharing of data in the cloud breaks the “silo” of hospital information, and imaging doctors can access and diagnose the original DICOM image data anytime and anywhere on the mobile terminal without being bound by time and location.
Ensure patient information security in the era of big data
Cloud computing has brought the security challenge of medical information while breaking the hospital information silo. The growing use of cloud-based software in healthcare, which allows health care organizations to share data with third parties through cloud-based software, including medical orders, electronic medical records (EMRs), medical images, cost details, etc., is a trend that poses significant security challenges for healthcare organizations, and this challenge is set to grow. Such detailed patient information makes medical data more valuable and attractive to hackers, and the consequences of such detailed medical data in the event of a hack are unthinkable. In addition, it is well known that “patient consent” is also very important in healthcare. “Many of our vendors that provide critical applications such as electronic medical records are moving aggressively to the cloud,” said John Houston, legal counsel at the University of Pittsburgh Medical Center, “and in many cases, we will have no choice. ”
In the era of big data, all patient medical data should be secured, and patients must have informed rights to third-party access to medical data. The HITRUST Cybersecurity Framework (CSF) is currently being introduced internationally, and the HITRUST CSF security program is the most widely recognized security certification in the healthcare industry. It combines existing regulations (such as HIPAA / HITECH, PCI, ISO 27001 and MARS-E) with industry best practices for healthcare-specific security, privacy and regulatory requirements.
Quality Control Software for Imaging Physicians
The budding imaging physician has limited experience and needs to improve his or her film review. Wouldn’t the diagnostic rate of imaging physicians be higher if there was an imaging physician quality control software? With this software, imaging trainees can repeatedly review the images they have already diagnosed and even receive review comments from senior imaging physicians. Even without an attending physician personally reviewing their work, junior imaging physicians can complete diagnostic imaging independently.
Ashimiyu B et al. from Johns Hopkins University School of Medicine, USA, used the Experience API (xAPI) framework to model a diagnostic radiology residency program in a region, which enables the acquisition, integration, and sharing of big data on learning behaviors based on the xAPI specification, laying the foundation for further analysis and mining of this diagnostic radiology residency program. xAPI provides A learner-centered model for capturing learning behavior data that supports learning record data from multiple data source systems, which is free from the traditional model of relying entirely on a single learning record collection, and is particularly suitable for mobile and Internet learning. The Graduate Medical Education Council (GMEC) recognizes this as a good way to assess residents and provide feedback on learning outcomes, and requires residency programs that implement interactive formative feedback assessments for residents, an assessment and feedback mechanism that allows imaging physicians to better diagnose and learn.
Patient-specific image interpretation software
To enable patients to participate in their medical care and decision-making, effective communication is critical, with clinical reports being the primary medium of communication. Effective patient-physician communication not only facilitates physicians to diagnose diseases, but also enhances the trust between doctors and patients. Increasingly, patients are accessing diagnostic imaging results or clinical data through electronic health record systems, but physicians often have less time to answer specific medical implications in detail for each patient. The University of Pennsylvania has developed a software system called PORTER that “translates” the medical terminology associated with radiology reports. If a patient does not know the terminology in the report, a mouse click on the term will explain the meaning of the term in easy-to-understand language so that the patient can better understand what the imaging physician is saying in the report. The software is HIPAA compliant and analyzes 100 knee MRI reports from academic medical centers to develop a glossary of 313 medical terms.
As the patient community relies more and more on data and cloud computing, it is believed that this innovative model will continue to adapt to this need, providing patients with more tangible information and more effective patient-physician communication.
Home Health Care
As global aging increases, the integration of home healthcare has become a major trend. Cuff blood pressure meter, intelligent blood glucose meter, digital stethoscope and other IoT medical devices can enable patients to manage hypertension, diabetes and other chronic diseases at home. The relevant test data can be transmitted to the doctor in real time through the cloud, and once there are data abnormalities, the doctor can also get in touch with the patient in a timely manner. In the future, “home hospitals” are likely to become more common in patients’ lives. If you’re a medical technology company or healthcare organization, this is a trend to watch and keep an eye on.
It is extremely challenging for patients and healthcare professionals to fully manage or prevent COPD-related hospitalizations in high-risk populations. To address this issue, Epharmix has developed EpxCOPD, a unique, data-based remote patient monitoring tool. Implemented by clinicians at Washington University School of Medicine and Saint Louis University School of Medicine, EpxCOPD tracks a patient’s daily breathing status to minimize preventable hospitalizations. EpxCOPD works in two ways: by sending relevant data, along with its medical phenotype, to the patient himself or by sending the data to the physician. This process allows for timely outreach and thus helps prevent hospitalizations for COPD patients.
ForaCare recently launched a smart glucose monitoring system called FORA GTel, an innovative product that will provide patients and caregivers with a unique and effective way to treat high-risk diabetes. FORA GTel provides glucose and blood ketone cell monitoring capabilities and allows for two-way communication between patients and physicians, providing an alternative to manual logging and downloading patient results. It also facilitates easy sharing of real-time data, keeping patients with diabetes and their care teams informed and engaged. The system also connects to FORA’s 24/7 HealthView telehealth service, which is HIPAA-compliant and allows healthcare professionals to monitor patients’ health status remotely.
“Sometimes healing, often caring, always comforting.” Many would quote Dr. Trudeau over 100 years ago to explain the nature of healthcare. Advances in technology will certainly cross the gap between doctors and patients, increase trust between them, and bring them closer together for a healthier life. I hope these six medical trends predicted by Forbes will benefit the people soon!