Description:BACKGROUND OF THE INVENTION

Field of Invention:

[0001] The present invention relates to the field of medical devices and, more specifically, to a hand-held device for brain stroke classification using deep learning and machine learning algorithms on computed tomography (CT) or Magnetic Resonance Imaging (MRI) printed on radiographic films.

Description of the Related Art

[0002] Stroke is a significant global health problem and a major cause of mortality and morbidity in developed countries and increasingly in low-middle-income countries (LMICs) [1]. A recent study has revealed that stroke affects people at a younger age in India compared to other parts of the world, with the average age of stroke onset being 55 years in India, as opposed to 70 years in Western countries [2]. This places a significant burden on India's healthcare system, where the number of neurologists is just 1200 for the world’s second-largest populous country [3].

[0003] Computed tomography (CT) and magnetic resonance imaging (MRI) are two commonly used diagnostic imaging techniques for the diagnosis of stroke in India and around the world. CT scans are often the first-line imaging modality for stroke diagnosis, as they are readily available and can quickly differentiate between ischemic and hemorrhagic stroke or normal. MRI scans can provide more detailed information about the brain tissue and blood vessels and are particularly useful for detecting small or subtle strokes that may be missed on a CT scan. There are a total of 3600 CT/MRI scanners in India.

[0004] In India, it is still common practice to print CT or MRI scans on radiographic film. This is often done for archival purposes, as physical films can provide a permanent record of the diagnostic image. Since the number of neurologists per population in India is abysmally low, an automated diagnostic system based on physical films can provide a first line of diagnosis. It has several benefits, such as faster and more accurate diagnosis of stroke, cost savings, and increased efficiency.

[0005] Cell phones have evolved in recent years from being straightforward specialized telecommunications devices to being compact, portable computers with the ability to carry out intricate, memory- and processor-intensive processes. These new smart gadgets, commonly referred to as smartphones, can offer an appealing means of offering image-based diagnostic services at a reasonable price due to advancements in medical imaging.

[0006] As such, the new generation of smart handheld devices with sophisticated hardware and operating systems has provided a portable platform for running medical diagnostic software, such as heart rate monitoring, diabetes monitoring, and experience sampling applications, which combine the usefulness of medical diagnosis with the convenience of a handheld device. Their light operating systems, such as the Apple®, iOS®, and Google® Android®, the support for user-friendly touch gestures, the availability of an SDK for fast application development, the rapid and regular improvements in hardware, and the availability of fast wireless networking over Wi-Fi and 3G make these devices ideal for medical applications.

SUMMARY OF THE INVENTION

[0007] The present invention is defined by the appended claims.

[0008] The present invention is directed to a portable imaging system for imaging the radiographic films. The portable imaging system comprises a hand holdable imaging device having a digital camera, a display, memory, a processor and a network connection and a library of algorithms tangibly stored in the memory and executable by the processor, where the algorithms are configured to classify the image printed on the radiographic film. The imaging of radiographic film could be complete or partial. Complete imaging would comprise the whole area of the radiographic film, and partial imaging would comprise part of the radiographic film.

[0009] The present invention is directed further to a digital processor-implemented system for classifying an image on radiographic film in real time. The system comprises a portable smart device comprising the processor, a memory, and a network connection and modules tangibly stored in the memory. The modules comprise a module for image normalization of an image object, a module for feature extraction within the normalized image, and a module for classification of the object based on extracted features.

[0010] The present invention is directed toward a related portable imaging system further comprising algorithms tangibly stored and processor-executable algorithms configured to display the image captured of a radiographic film and the results of the classification thereof.

[0011] The present invention is directed toward a related portable imaging system further comprising algorithms tangibly stored and processor-executable algorithms configured to generate a report comprising of the captured image, processed image displaying the approximate area of the image on the basis of which decision has been taken by the algorithm and results of classification of the captured image.

[0012] The present invention is directed toward a related portable imaging system further comprising algorithms tangibly stored and processor-executable algorithms configured to transmit the generated report over the communication network (secured or unsecured) to other devices.
, C , Claims:Claims

1. A digital processor-implemented system for classifying a radiographic film, comprising:

a portable smart device comprising a digital camera, the processor, a memory and a network connection; and

modules tangibly stored in the memory comprising:
A module for data preprocessing (100)
a module for feature extraction (200) within the segmented images;
a module for classification (300) of the image based on extracted features; and
a module for display of the object of interest and results of the classification thereof;

2. A digital processor-implemented system for calculating patient-level classification on the basis of a series of images obtained from the radiographic film.

3. A digital processor-implemented system for displaying the results of the classification and sharing of generated reports over a network communication channel.