초록 ▼

The present invention provides devices for screening the skin of an individual in real time using a region-fusion based segmentation with narrow band graph partitioning algorithm to analyze and classify a region of interest of the skin as benign or malignant. Also, provided is a method for screening

The present invention provides devices for screening the skin of an individual in real time using a region-fusion based segmentation with narrow band graph partitioning algorithm to analyze and classify a region of interest of the skin as benign or malignant. Also, provided is a method for screening the skin of an individual using the devices described herein. In addition the present invention provides a digital processor-implemented system for classifying a region of interest on the skin and a processor readable medium having processor-executable instructions to perform skin cancer detection.

대표청구항 ▼

1. A device for screening the skin of an individual in real time, comprising: in a self-contained, hand-held and portable means for housing the device: means for programming the device;means for controlling interaction with a user of the device;means for acquiring and displaying an image of a region

1. A device for screening the skin of an individual in real time, comprising: in a self-contained, hand-held and portable means for housing the device: means for programming the device;means for controlling interaction with a user of the device;means for acquiring and displaying an image of a region of interest on the skin of an individual;means for analyzing and classifying the image comprising a region-fusion based segmentation with narrow band graph partitioning algorithm with an active contour model; andmeans for controlling one or both of an exchange of data with a personal computer or a receipt of software from a personal computer. 2. The device of claim 1, further comprising means for powering the device. 3. The device of claim 2, wherein the means for powering the device comprises a power management unit having a battery, a battery charger, a supervisor and voltage regulators. 4. The device of claim 1, wherein the means for programming the device comprises: a programmable digital signal microprocessor with a memory storing processor executable instructions; and a boot flash for parallel flash memory. 5. The device of claim 1, wherein the means for controlling interaction with a user of the device comprises: a human machine interface board, including a keypad and one or more display terminals; and a user interface. 6. The device of claim 5, wherein the display terminals comprise one or both of an on-board display terminal or an off-board display terminal. 7. The device of claim 1, wherein the means for acquiring and displaying the image of the skin comprises: a sensor having a combination of one or more light sources and one or more imaging sensors for image acquisition; a video encoder to convert the acquired digital image as an analog off board image; and an on board visual display. 8. The device of claim 1, wherein the means for controlling one or both of an exchange of data with a personal computer or receipt of software from a personal computer comprises a device configured to wirelessly communicate between a programmable digital signal microprocessor comprising the handheld screening device and a personal computer, including a parallel for a Universal Asynchronous Receiver/Transmitter. 9. A device for screening a region of interest on the skin of an individual in real time, comprising: in a self-contained, hand-held, portable housing, a programmable digital signal microprocessor with a memory storing processor executable instructions digitally linked to: a boot flash for parallel flash memory digitally connected to the microprocessor;an input device;a sensor having a combination of one or more light sources and one or more imaging sensors;a digital to analog signal video encoder connected to an off-board display terminal;an on-board display terminal;a wireless device configured to wirelessly communicate between the programmable digital signal microprocessor and a personal computer, including a parallel for a Universal Asynchronous Receiver/Transmitter;a power management unit;a module for pre-processing and image segmentation of the region of interest comprising region-fusion based segmentation with narrow band graph partitioning algorithm with an active contour model;a module for image classification of a lesion found within the region of interest as a benign or malignant lesion as determined by one or more of a size difference, shape irregularity or texture formation each of which comprises a partial classification;a module for decision voting in terms of confidence based weighted averaging of the partial classifications; anda module for image reconstruction for three-dimensional imaging of the region of interest. 10. The device of claim 9, wherein the input device is a keypad or a touch screen. 11. The device of claim 9, wherein the light source(s) emit visible or invisible light arranged in different spatial configurations and wherein the imaging sensor(s) is an optical camera, an infrared camera, a thermal camera, a three dimensional stereo camera or a digital light processing sensor. 12. The device of claim 9, wherein the power management unit comprises a battery, battery charger, supervisor, and two voltage regulators. 13. A method for screening the skin of an individual in real time, comprising: acquiring one or more images of a region of interest on the skin in one or more modalities with the device of claim 1;displaying the acquired image;preprocessing the acquired image(s) using the region-fusion based segmentation with narrow band graph partitioning algorithm with an active contour model;classifying the region of interest on the skin as benign or malignant; anddisplaying the classification results. 14. The method of claim 13, further comprising: wirelessly communicating the classification results to a personal computer. 15. The method of claim 13, wherein the imaging modality is XLM or TLM. 16. The method of claim 13, wherein preprocessing comprises extracting a lesion from the background of the image and determining a lesion border. 17. The method of claim 13, wherein classifying the lesion comprises confidence based weighted averaging of a combination of partial classifications based on one or more features of the lesion. 18. The method of claim 17, wherein the features of the skin lesion include a size difference, shape irregularity or texture formation. 19. The method of claim 13, wherein displaying the classification comprises color coding the results. 20. A digital processor-implemented system for classifying a region of interest on the skin of an individual, comprising: a module for pre-processing and image segmentation of the region of interest comprising a region-fusion based segmentation with narrow band graph partitioning algorithm with an active contour model;a module for image classification of a lesion found within the region of interest as a benign or malignant lesion as determined by one or more of a size difference, shape irregularity or texture formation each of which comprises a partial classification;a module for decision voting in terms of confidence based weighted averaging of the partial classifications; anda module for image reconstruction for three-dimensional imaging of the region of interest. 21. The processor implemented system of claim 20, further comprising a module for wireless communication between the processor and a personal computer. 22. A processor readable medium having processor-executable instructions to perform a method for detecting skin cancers on an individual in real time, said method comprising the steps of: acquiring one or more images of a region of interest on the skin in one or more modalities;preprocessing the acquired image(s) comprising extracting a lesion from the background of the image and determining a lesion border using a region-fusion based segmentation with narrow band graph partitioning algorithm with an active contour model;classifying the skin lesion as benign or malignant by a confidence based weighted averaging of a combination of partial classifications based on one or more features of the lesion; anddisplaying the classification results. 23. The processor readable medium of claim 22, wherein the imaging modalities are XLM or TLM. 24. The processor readable medium of claim 22, wherein the features of the lesion include a size difference, shape irregularity or texture formation. 25. The processor readable medium of claim 22, wherein displaying the classification comprises color coding the results.