Tiangong University to Transfer Brain Hemorrhage MIT Detection Patent for RMB 600,000
Recently, Tiangong University released a public notice on the transformation of scientific and technological achievements, proposing to transfer its“A Method, Apparatus, Device, Medium, and Product for Detecting MIT Regions in Cerebral Hemorrhage”Transferred to Tianjin Baiwangda Technology Co., Ltd. via patent assignment, with a proposed transaction price ofRMB 600,000。
This patent is held byChen Ruijuan, Zhu Xinlei, Wang HuiquanJointly developed by three parties, with the core focus onApplication of Magnetic Induction Tomography (MIT) Technology in the Detection of Intracerebral HemorrhageThis patent addresses the limitations of MIT systems, such as limited imaging information and susceptibility to noise, through technical designs involving regional partitioning, model training, and targeted detection. As a result, it enhances the image reconstruction quality and accuracy for detecting cerebral hemorrhage regions, providing a low-cost, radiation-free novel auxiliary detection solution for preoperative, intraoperative, and postoperative monitoring of cerebral hemorrhage.
In the field of clinical detection for intracerebral hemorrhage, although mainstream imaging modalities such as CT and MRI offer superior image quality, they fail to meet the clinical needs for real-time intraoperative and preoperative monitoring, as well as low-cost adjunctive testing.Magnetic Induction Tomography (MIT)As a novel, non-contact, radiation-free imaging technology, it possesses inherent advantages such as low cost and real-time monitoring capability, making it a high-quality option for assisting in the detection of intracerebral hemorrhage; however, it faces significant technical challenges in practical commercial applications.
On the one hand, the construction of Magnetic Induction Tomography (MIT) systems is constrained by spatial limitations. To ensure ease of use, there is a clear upper limit on the number of coils that can be configured, which directly restricts the imaging information obtainable by the device. This makes it difficult to support high-precision lesion detection and image reconstruction. On the other hand, during actual clinical application, MIT detection signals are highly susceptible to environmental noise interference, significantly compromising signal stability and accuracy, thereby further degrading imaging quality and detection precision.
These two major pain points have become key factors constraining the large-scale and commercial application of MIT technology in the field of cerebral hemorrhage detection, making it difficult for this technology to fully realize its auxiliary value in preoperative, intraoperative, and postoperative monitoring of cerebral hemorrhage.
To address the above two major pain points, this intracerebral hemorrhage MIT region detection technology focuses onEnhancing Imaging AccuracywithNoise ImmunityCore objectives have driven multidimensional technological innovation, with key innovations manifested in four major areas: detection methods, model construction, device adaptation, and signal utilization.
First, an innovative coil array adaptation strategy based on zonal detection is proposed.The brain imaging region is divided into four 90° sectors, with dedicated coil arrays configured for each sector. The excitation and detection coil arrays are positioned 180° apart relative to the center of the patient’s brain and can be rotated and adjusted to predefined angles, ensuring that the lesion remains within the sensitive detection zone ranging from -45° to +45°. This approach addresses the issue of insufficient imaging information caused by a limited number of coils and enhances the prominence of lesion signal characteristics.
Second, a customized MIT prediction model system was constructed.Magnetic field and conductivity data acquired from different coil arrays are used as sample pairs to construct dedicated datasets. Stacked autoencoders are trained separately to generate multiple prediction models, establishing a one-to-one correspondence between each coil array and its respective prediction model. This approach overcomes the limitations of poor adaptability associated with traditional single-model methods, enabling more targeted interpretation of detection data.
Third, integrating multiple technologies to achieve precise signal processing and imaging.First, the lesion location is determined using established modalities such as CT and MRI. Then, a lock-in amplifier is employed to convert magnetic field information acquired by the detection coil into phase information. Finally, image reconstruction is performed with phase information as the input and conductivity distribution as the output. This approach achieves complementary advantages of different detection techniques, thereby improving the accuracy of lesion localization and imaging.
Fourth, innovatively applying the Stacked Autoencoder (SAE) neural network algorithm for image reconstruction。By integrating the physical principles of MIT technology, the algorithm is fundamentally based on the linear relationship between phase difference and conductor conductivity. This approach not only fully extracts lesion location information from detection signals but also significantly enhances the technology's noise immunity, enabling high-quality lesion reconstruction even under low-noise conditions.
Currently, there are few commercially available products for intracerebral hemorrhage detection based on magnetic induction technology worldwide. Hangzhou Yongchuan Technology’s MH-200 and Cerebrotech’s Visor series (CMS-5000) from the United States are representative products in this field. Both leverage magnetic induction technology to enable non-invasive detection of intracerebral hemorrhage, focusing on core clinical scenarios such as pre-hospital emergency care and bedside monitoring, thereby serving as typical examples of the clinical application of Magnetic Induction Tomography (MIT) technology in intracerebral hemorrhage detection.
Hangzhou Yongchuan Technology's MH-200 Electrical Impedance Tomography SystemIt is the first magnetic induction electrical impedance tomography (MIEIT)-based cerebral monitoring device to obtain relevant registration certification in China. Leveraging core MIT technology, it enables non-contact measurement of cerebral electrical impedance distribution, allowing for the identification of intracerebral hemorrhage lesions and non-invasive, real-time monitoring of disease progression. The device provides clinicians with visualized imaging and diagnostic/therapeutic indicators. With strong adaptability, it has already been deployed in various medical institutions across China, particularly in emergency departments and pre-hospital emergency care settings.
U.S. Cerebrotech's Visor Series (CMS-5000)It is a portable, non-invasive cerebral monitoring device approved by the U.S. FDA. Based on volumetric impedance phase-shift spectroscopy technology, it identifies intracranial fluid abnormalities caused by cerebral lesions such as intracerebral hemorrhage and cerebral edema by detecting changes in intracranial bioimpedance. The device offers convenient operation and rapid feedback, providing objective data to assist clinicians in determining the type of stroke. It is currently being used in pre-hospital emergency care and hospital emergency departments in several countries across Europe and the United States.
These products and innovations not only provide new perspectives for technological upgrades in the field of intracerebral hemorrhage detection, but also offer more cost-effective diagnostic solutions for scenarios such as primary healthcare and clinical monitoring, thereby promoting diversified development in the field of neurological medical imaging.