Much of the technology presented this year MEDICA 2022 relates to the field of radiology and medical imaging. MedicalExpo e-magazine caught up with Takashi Azuma, CTO of a Japanese ultrasound imaging company Lily MedTechand Sophia Borowka, CEO of the Swiss ultrasound device company aiSon Technologiesto discuss some of the latest radiology trends and technologies that will be showcased at MEDICA 2022. Both companies are participating in the event.
1/ Ultrasound computed tomography
Many people are familiar with common medical imaging techniques such as X-ray tomography or magnetic resonance imaging (MRI). Most people are also familiar with b-mode ultrasound (grayscale imaging), which is typically used to produce images of growing fetuses.
A promising alternative is ultrasound computed tomography (USCT), which uses high-frequency sound waves emitted along a 2D slice of the human body to image interior organs and bones. USCT is less expensive than MRI and there is no ionizing radiation involved, as there is with X-ray tomography.
Takashi Azuma said:
“Ultrasound imaging has grown significantly in terms of portability and ease of use in recent years. Attempts to use it in new areas, such as orthopedics, are now expanding beyond conventional diagnostic targets.
Difficulties in controlling the accuracy of ultrasound imaging have been resolved with the evolution of USCT, which enables non-contact 3D imaging and records the full range needed. Currently, breast cancer remains the primary target, but joint imaging and new areas such as the brain are also being explored.
In 2021, Lily Medtech launched COCOLYan innovative bed-type ultrasound examination device that uses a ring ultrasound transducer to check for breast cancer.
2/ Portable ultrasound
Another significant trend in ultrasound imaging is the shift to portable devices, spurred by advances in technology and the transition from hospital to outpatient care. Today, the portable ultrasound market is growing rapidly as the latest generation of ultra-portable devices are increasingly accepted by healthcare professionals.
Sophia Borowka said:
“Thanks to intense research and technological advances, the image quality delivered by these devices continues to improve. Even superficial soft tissue structures, where ultrasound can trump CT and MRI imaging in resolution and predictivity, can be easily imaged with specialized handheld ultrasound devices.
As portable ultrasound technology has evolved, so have spreader pads. By increasing the distance between the ultrasound transducer and the patient’s skin, such electrodes make it possible to image superficial lesions, lesions immediately under the skin, superficial foreign bodies and breast lesions. She added:
“aiSon Technologies has now developed the first adaptive ultrasound cushion. These reduce the amount of pressure that must be exerted on the patient’s skin, allowing an unobstructed view of the higher anatomical structures. This avoids misdiagnosis in rheumatology and greatly facilitates ultrasound-guided hand surgery.
3/ New generation CADe and CADx
Traditional computer-aided detection (CAD) in mammography is nothing new – computers have been used to improve mammographic abnormality detection since the 1960s. Yet CAD has taken on a broader meaning more recently, especially in the context of renewed interest in artificial intelligence (IA).
Over the past decade, significant research has been conducted on deep learning. This has been made possible by advanced algorithms powered by faster computers, greater storage capacities and the availability of big data.
New deep learning-based mammography platforms, which differ from traditional CAD in several important ways, are expected to significantly improve radiological accuracy. Two components should be considered: computer-aided detection (CADe), which focuses on detection, and computer-aided diagnosis (CADx), which focuses on classification. Compared to traditional CAD, next-generation systems will provide both CADe and CADx, providing a fully automated end-to-end process and better facilitating machine learning.
Takashi Azuma said:
“The development of AI, which bridges the gap between the capabilities of non-specialists and specialists, has influenced many imaging modalities and expanded the number of target diseases.
In the future, AI has the potential to learn not only from images of patients with lesions, but also images from previous examinations, enabling the discovery of lesions that even trained specialists cannot find.
Combined with the development of image-guided liquid biopsies and minimally invasive treatments, AI promises to revolutionize the detection, diagnosis and elimination of cancer.