Method and system of analyzing medical images

#The present invention relates to analyzing images; particularly, the present invention relates to analyzing medical images.

#The number of hip fractures and the resulting post-surgical outcome are significant public health concerns worldwide. The number of hip fractures continues to increase due to prolonged human lifespan and the growing elderly populations. Approximately 20-30% of patients with a hip fracture encounter life-threatening situations within a year, and the majority of the patients experience significant functional loss. Because of the weight-bearing nature of the region, patients who have clinically ‘silent’ fractures rapidly develop severe pain and immobility. Early detection and surgery are critical for patient survival and the preservation of hip function. Postponed management of hip fractures results in a poor prognosis and even an increased risk of death years later. Therefore, detecting hip fractures as soon as possible is critical for remote mortality and medical outcomes.

#Frontal pelvic radiographs (PXRs) are an essential and widely used tool for image evaluation for hip fractures. However, hip fracture assessment using frontal pelvic radiographs is not optimal. Studies show that mis-diagnosis by qualified medical professionals, are as high as 7-14%. The mis-diagnosis delays the proper diagnosis and treatments, and worsen prognosis of hip fracture. To avoid further health sequelae and medical costs associated with a delayed diagnosis, additional radiographs, nuclear medicine bone scans, computed tomography (CT) scans, and magnetic resonance imaging (MRI) scans have been recommended as routine diagnostics. However, it is not an effective, efficient or economical method to use these diagnostic tools as routine examinations.

#FIG. 1 is a schematic diagram of an embodiment of a system of analyzing medical image;

#FIG. 2 is a flow chart of an embodiment of a method of analyzing medical image;

#FIG. 3 is an assembly of images wherein images a, c and e are original pelvic radiographs with arrows indicating the fracture; and images b, d and f are Grad-CAM images generated according to an embodiment of the present invention corresponding to images a, c and e, respectively.

#An embodiment of the present invention is described with reference to FIG. 1.

#A system 100 of analyzing medical image is provided. The system 100 comprises a scanner 110 for receiving a medical image. The scanner may be an X-ray (radiograph), Magnetic resonance imaging (MRI) or a computed tomography (CT) scanner. The medical image may be X-ray, CT or MRI image of a region of a subject (such as an upper/lower body of a subject) or a specific-site of a subject. The system further comprises a processor 120 for (a) analyzing the medical image, (b) determining if the medical image includes a presence of fracture and (c) transmitting an indication based on the determination in step (b) to a display 130 for visually presenting the indication. The processor may be a microprocessor or a processing unit, such as a field-programmable gate array (FPGA) or other suitable programmable processing unit. The display 130 may be a screen of a computer or a smartphone, a light emitting diode (LED) light indicator or any other suitable display capable of visually presenting the indication.

#Referring to FIG. 2, when in operation, the processor 120 receives a medical image at step 200. The medical image can be a radiograph such as an X-ray, MRI or a CT scan. At step 210, the processor 120 applies a model stored in a memory. The memory can be a storage of the processor 120 or a storage external to and in communication with the system 100. The model may be an iteratively learning model, or a model generated from pre-training processes. The system analyzes the medical image based on the model at step 220 and determines if the medical image includes a presence of fracture at step 230. At step 240, the processor 120 transmits an indication indicative of the determination.