Vesta Teleradiology | artificial intelligence

March AI News in Diagnostic Imaging

New Research by Harvard Medical School, MIT and Stanford on AI and Clinician Performance

The potential of medical artificial intelligence (AI) tools to enhance clinicians’ performance in interpreting medical images varies among individual clinicians, as highlighted by recent research led by Harvard Medical School, MIT, and Stanford. Published in Nature Medicine, the study underscores the intricate nature of human-AI interaction, which remains incompletely understood. While some radiologists benefit from AI assistance, others experience interference, affecting diagnostic accuracy.

The findings stress the necessity for personalized AI systems tailored to individual clinicians, emphasizing careful implementation to maximize benefits and minimize harm. Despite variations in AI’s impact, the results shouldn’t deter AI adoption but rather prompt a deeper understanding of human-AI dynamics to design approaches that enhance human performance.

To ensure effective integration of AI in clinical practice, collaboration between AI developers and clinicians is essential, alongside rigorous testing in real-world scenarios. Furthermore, efforts should focus on improving AI accuracy and training radiologists to discern AI inaccuracies, facilitating informed decision-making. Ultimately, understanding the complexities of machine-human interaction is pivotal for optimizing patient care through AI integration in radiology.

AI and Workflows

New research highlights a novel reporting workflow that automatically incorporates artificial intelligence (AI) findings into structured radiology reports, streamlining physicians’ tasks and saving valuable time. German experts shared their experience with the “AI to SR pipeline,” which integrates a commercially available AI tool for chest X-ray pathology detection and localization into structured report templates.

In evaluations conducted at University Medical Center Mainz, expert radiologists found that reports generated using the AI to SR pipeline were faster compared to free-text reporting and conventional structured reporting. Additionally, subjective quality assessments indicated higher ratings for reports created with the pipeline.

In the hospital’s clinical routine, chest X-ray images are sent to the picture archiving and communication system, then automatically forwarded to the AI tool for analysis. The results are output in a DICOM structured reporting format, taking approximately five minutes from image acquisition to final reporting. Radiologists were able to create chest X-ray reports significantly faster with the pipeline compared to free-text and conventional structured reporting, while also rating the AI-generated reports more favorably.

The authors suggest that this AI-driven reporting pipeline offers standardized, time-efficient, and high-quality reporting for chest X-rays, potentially enhancing AI integration into daily clinical practice and maximizing its benefits.

Sources:

Medicalxpress.com
Radiologybusiness.com
Openai.com

The Latest in Brain Imaging News

In recent years, awareness surrounding brain injuries has steadily risen, prompting significant strides in diagnostic technologies and treatment modalities. As we delve into the latest developments in this critical area of healthcare, it becomes increasingly apparent that advancements in medical imaging, particularly in the realm of neurological disorders, are poised to revolutionize the landscape of brain injury diagnosis and management.

AI-based Quantitative Brain Imaging System

Philips and Synthetic MR have joined forces to advance the diagnosis of neurological disorders through cutting-edge quantitative brain imaging tools. Their collaboration introduces the Smart Quant Neuro 3D MRI software suite, combining Philips’ SmartSpeed image-reconstruction technology, the 3D SyntAc clinical application, and SyntheticMR’s SyMRI NEURO 3D software. This innovation employs AI to analyze brain tissues, enhancing the detection and analysis of conditions like multiple sclerosis, traumatic brain injuries, and dementia.

The rise of AI in diagnostic imaging, projected to reach $1.2bn by 2027, signifies a transformative shift in improving accuracy and patient outcomes. With the diagnostic imaging market expected to grow to $9.1bn by 2030, fueled by demand for early disease diagnosis and personalized medicine, this partnership underscores the crucial role of AI in enhancing medical imaging.

Read the press release here.

A New Way of Diagnosing Mild TBIs

Researchers have developed a novel brain imaging method to diagnose mild traumatic brain injuries (mTBIs), which are often missed by standard techniques like MRI. This method involves loading gadolinium, a common MRI contrast agent, into micropatches attached to immune cells called macrophages. These cells migrate to areas of brain inflammation caused by mTBIs, enabling MRI detection. The technique, called M-GLAMs, was successfully tested in mice and pigs, showing promise for accurately diagnosing mTBIs. It also allows imaging at lower gadolinium doses, potentially benefiting patients with kidney issues. While unable to pinpoint injury locations, M-GLAMs could aid in identifying and treating brain inflammation. The researchers aim to bring this technology to clinical trials, with support from grants and intellectual property protection.

Read the study here.

New Imaging Tech that Captures Neuronal Activity Across the Brain During Recovery

Researchers at Tufts University School of Medicine have developed a novel imaging technology to monitor neuronal activity throughout the entire brain during the initial weeks of recovery from traumatic brain injury (TBI). Their study, published in Cerebral Cortex, reveals that TBI can induce changes in brain function beyond the injury site. Using a combination of fluorescent sensors and electrodes, they observed altered connectivity patterns in mice post-injury, even in regions distant from the impact. Despite the mice’s ability to perform physical tasks normally, their brain activity during both exercise and rest differed significantly from healthy brains. This impaired ability to switch between states suggests underlying brain state dysfunction post-injury. The findings highlight the brain’s plasticity in response to injury and have potential clinical implications for understanding TBI impacts and tailoring treatments. The researchers aim to further investigate long-term neural activity changes post-recovery and explore the technology’s potential in predicting specific dysfunctions or long-term outcomes of TBI.

Read the study here.

Sources:

Medicaldevice-network.com
Otd.harvard.edu
Scitechdaily.com
Openai.com

Vesta Teleradiology Partners with MIT for AI Research

Artificial intelligence is a young field of study that has been growing exponentially as experts try to identify ways to integrate the help of machines to solve human problems. Computer systems have great potential to become facilitators in helping detect cancers and other medical conditions. It is clear that in order to advance our capabilities in patient care, we must embrace AI as it is reshaping the healthcare industry.

Massachusetts Institute of Technology (MIT) has been a pillar in research for the sciences and has been known to provide breakthroughs that benefit our lives.

MIT is in the process of developing an innovative AI Radiology application that will assist radiologists in improving the quality and speed of interpretations. Vesta Teleradiology is honored to partner with the research team in assisting with the development of this project.

Vesta has been working with various universities in supporting their needs and this partnership with MIT is inline with Vesta’s goal of helping further the research in the industry.

Each of the highly skilled Vesta radiologists will be interpreting a batch of randomly selected studies using a pre-set criteria. The interpretations are used to help enhance the algorithm and help its quality in the process. Vesta and MIT teams work very closely to make the project a success. It is through this partnership that Vesta hopes to considerably advance the field of AI in radiology and imaging.

Teleradiology Company

Vesta Teleradiology provides preliminary and final interpretations for imaging studies. They believe it is crucial to incorporate new technologies into their offerings to provide efficient solutions for healthcare providers, from hospitals to private physicians and even universities. Being at the forefront of medical advancements is part of the Vesta philosophy.

MIT is a technological innovator and is leading this effort in radiology. The contract between Vesta and MIT is sure to be one to help advance the future of medicine.

About Vesta:

Vesta exists to make life better for their healthcare facility clients and their patients through efficient teleradiology services and seamless systems integrations.

How AI is Making an Impact on Radiology and Imaging

The fields of science and medicine are always progressing. This progression intends to help both patients and providers.

Today, artificial intelligence (AI) is becoming common as a way to diagnose patients. It provides a more efficient way to collect and store information. The software can even analyze imaging to a high level of accuracy. This helps providers catch a problem that they may have missed before.

AI is a field that is advancing quickly. What progress have we seen in the past couple of years? What programs have we begun to put in place?

What Is Artificial Intelligence?

Artificial intelligence refers to highly advanced computers or computer-controlled robots. These computers are capable of performing incredibly complex tasks. Before, we thought these tasks could only be done by intelligent beings.

AI in imaging — AI is making advancements in the medical field

These computers are often associated with human characteristics. They seem to be able to reason and learn from past experiences.

How Is Artificial Intelligence Used For Diagnostic Imaging & Radiation?

Using AI in radiology and imaging has been gaining traction in the medical world. We use it largely to store and analyze data, helping physicians to make a prognosis. AI can store and analyze all a patient’s records. It can then make a diagnosis based on those records. The analysis is often far more accurate than what a human counterpart can do.

The use of AI is also helpful because of its storage capability. AI can have large imaging biobanks to hold more images than standard computers.

It also makes the lives of physicians easier by filtering patients by need. It can recommend appropriate diagnostic imaging based on the patient’s current records. It can also sort patients by priority in the case of an emergency.

What Advancements Have Been Made?

AI means to eliminate problems associated with human limitations. Traditional imaging takes a team of technicians. They must take the imaging as well as interpret it. This can be time-consuming. Plus, AI is able to analyze images with far greater accuracy than the human eye.

Radiomics

Radiomics is a tool that performs a deep analysis of tumors down to the molecular level. AI can perform radiomics with far better accuracy than the human eye or brain.

AI can analyze a specific region and extract over 400 elements. It then takes these features and correlates them with other data to form a diagnosis. The AI can analyze features from radiographs, CT, MRI, or PET studies.

Rapid Brain-Imaging AI Software

Hyperfine is the manufacturer of portable MRI machines. They are now creating these machines with new AI intelligence software. They believe that this new software will be able to perform brain scans in under 3 minutes.

AI-Generated Drugs

In 2020, an AI-created drug went to human clinical trials. The drug intends to treat OCD, and was designed entirely by AI. Exscientia is the manufacturer of the drug. They say that it normally takes about 4.5 years to get a new drug to this stage of testing. With AI generation, the drug got to the human clinical trial stage in under 12 months.

Making A Diagnosis

We stated earlier that AI is being used as a way to more efficiently diagnose patients. Still, relying entirely on AI to do this can complicate things and may be unwise.

So, the researchers of MIT’s Computer Science and Artificial Intelligence Lab worked to combat this. They created a machine learning system that analyzes the data and decides whether to diagnose.

If it “feels” it’s unable to make an accurate prediction, it will defer to a medical professional. It even considers whether to defer to an expert based on who in the medical team is available. It will consider each team member’s availability, level of experience, and specialty.

Conclusion

AI in diagnostic imaging shows promise to truly advance quality of care for patients. We are excited to see more advancements in this arena. In the meantime, we don’t believe any machine can currently replace a trained human eye when it comes to interpretations. At Vesta, we provide US Board Certified radiologists who work to provide accurate preliminary and final interpretations. Learn how we can support your radiology department– contact us today.