# US 2014/0193051 A1 · Lesion diagnosis apparatus and method > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2014/0193051 A1 - Document ID: US20140193051A1 - Status: Published application - Kind: A1 - Application number: 14/151,168 - Filed: 2014-01-09 - Published: 2014-07-10 - Earliest claimed: 2013-01-10 - Assignees: Samsung Electronics Co., Ltd. - Applicants: Samsung Electronics Co., Ltd. - Inventors: Jae-Cheol LEE; Kyoung-Gu WOO; Yeong-Kyeong SEONG - Web page: https://patentgrep.com/patent/US20140193051A1 ## CPC - G06T 7/0012: {Biomedical image inspection} - G06T 2207/10016: Video; Image sequence - G06T 2207/10132: Ultrasound image - G06T 2207/20076: Probabilistic image processing - G06T 2207/20081: Training; Learning - G06T 2207/30096: Tumor; Lesion - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract A lesion diagnosis apparatus and a lesion diagnosis method are provided. A lesion-surrounding area determination unit is configured to determine an existence of a lesion-surrounding area from continuous medical images. A feedback provision unit is configured to generate feedback information about a presence of a lesion in the lesion-surrounding area. ## Claims ### Claim 1 A lesion diagnosis apparatus comprising: a lesion-surrounding area determination unit configured to determine an existence of a lesion-surrounding area from continuous medical images; and a feedback provision unit configured to generate feedback information about a presence of a lesion in the lesion-surrounding area. ### Claim 2 (depends on 1) The lesion diagnosis apparatus of claim 1, wherein the lesion-surrounding area determination unit determines the existence of the lesion-surrounding area from each of the continuous medical images. ### Claim 3 (depends on 1) The lesion diagnosis apparatus of claim 1, wherein the lesion-surrounding area determination unit determines the lesion-surrounding area from the continuous medical images using a discriminant model. ### Claim 4 (depends on 3) The lesion diagnosis apparatus of claim 3, wherein the discriminant model is generated by learning at least one of anatomical features and image features of a previously collected lesion-surrounding area. ### Claim 5 (depends on 1) The lesion diagnosis apparatus of claim 1, wherein the lesion-surrounding area determination unit determines whether there is a lesion-surrounding area by analyzing each frame of the continuously-received medical images or by sampling each of the continuously-received medical images every number of frames according to a selection of a user or a processing performance. ### Claim 6 (depends on 3) The lesion diagnosis apparatus of claim 3, wherein the lesion-surrounding area determination unit determines a possibility of a presence of a lesion as a numerical value by comparing a brightness variation of each medical image of the continuous medical images against the discriminant model. ### Claim 7 (depends on 1) The lesion diagnosis apparatus of claim 1, wherein the lesion-surrounding area determination unit determines a possibility of a presence of the lesion in the lesion-surrounding area from the continuous medical images according to a measured location of a medical image, and the feedback provision unit generates an image indicating the possibility of the presence of the lesion according to the location of the medical image. ### Claim 8 (depends on 7) The lesion diagnosis apparatus of claim 7, wherein the feedback provision unit generates, based on the possibility of the presence of the lesion, the image indicating a direction in which the lesion is likely to be present. ### Claim 9 (depends on 7) The lesion diagnosis apparatus of claim 7, wherein the feedback provision unit generates, based on the possibility of the presence of the lesion, the image in the form of probability map indicating the possibility of the presence of the lesion. ### Claim 10 (depends on 7) The lesion diagnosis apparatus of claim 7, wherein the feedback provision unit generates at least one of a visual signal, an audio signal, and a tactile signal to guide a measuring device toward a direction in which the lesion is likely to be present. ### Claim 11 (depends on 1) The lesion diagnosis apparatus of claim 1, further comprising: a diagnosis unit configured to diagnose the lesion from each of the continuous medical images; and a storage unit configured to store at least one of location information, features, and diagnosis of the lesion diagnosed, wherein the feedback provision unit further generates the feedback information to output the diagnosis of the diagnosed lesion. ### Claim 12 (depends on 11) The lesion diagnosis apparatus of claim 11, wherein the feedback provision unit further generates, based on at least one of the location information, the features, and the diagnosis, visual information to notify that the lesion has been diagnosed. ### Claim 13 (depends on 12) The lesion diagnosis apparatus of claim 12, wherein the diagnosis unit diagnoses, based on at least one of the location information and the features, an undiagnosed lesion. ### Claim 14 (depends on 11) The lesion diagnosis apparatus of claim 11, wherein the diagnosis unit diagnoses in real-time by detecting a lesion from at least one of or each of the continuously-received medical images, extracting features from the detected lesion, and classifying the detected lesion. ### Claim 15 (depends on 11) The lesion diagnosis apparatus of claim 11, wherein the diagnosis unit diagnoses the lesion based on all frames of each of the continuous medical images or based on sampling the continuously-received medical images at predefined number of frames. ### Claim 16 (depends on 11) The lesion diagnosis apparatus of claim 11, wherein the diagnosis unit determines whether the lesion is identical to a previously diagnosed lesion based on at least one of a relative location or absolute coordinates of location information of the lesion, and features of the lesion. ### Claim 17 (depends on 16) The lesion diagnosis apparatus of claim 16, wherein when the diagnosis unit determines that the lesion is identical to a previously diagnosed lesion, the diagnosis unit excludes diagnosing the lesion, wherein the features of the lesion comprise a boundary shape and boundary sharpness, an echo pattern, and posterior acoustic shadow. ### Claim 18 A lesion diagnosis method comprising: receiving continuous medical images; determining an existence of a lesion-surrounding area from the continuous medical images; and generating feedback information about a presence of a lesion in the lesion-surrounding area. ### Claim 19 (depends on 18) The lesion diagnosis method of claim 18, wherein the determining of the existence of the lesion-surrounding area is from each of the continuous medical images. ### Claim 20 (depends on 18) The lesion diagnosis method of claim 18, wherein the determining comprises determining the lesion-surrounding area using a discriminant model. ### Claim 21 (depends on 20) The lesion diagnosis method of claim 20, further comprising: generating the discriminant model by learning at least one of anatomical features and image features of a previously collected lesion-surrounding area. ### Claim 22 (depends on 18) The lesion diagnosis method of claim 18, wherein the determining comprises determining a possibility of a presence of the lesion in the lesion-surrounding area from the continuous medical images according to a measured location of a medical image, and the generating comprises generating an image indicating a possibility of the presence of the lesion according to the location of the medical image. ### Claim 23 (depends on 22) The lesion diagnosis method of claim 22, wherein the generating comprises, based on the possibility of the presence of the lesion, generating the image indicating a direction in which the lesion is likely to be present. ### Claim 24 (depends on 22) The lesion diagnosis method of claim 22, wherein the generating comprises, based on the possibility of the presence of the lesion, generating the image in a form of probability map indicating the possibility of the presence of the lesion. ### Claim 25 (depends on 22) The lesion diagnosis method of claim 22, wherein the generating comprises, based on the possibility of a presence of a lesion, generating at least one of a visual signal, an audio signal, and a tactile signal to guide a measuring device toward a direction in which the lesion is likely to be present. ### Claim 26 (depends on 18) The lesion diagnosis method of claim 18, further comprising: diagnosing the lesion from each of the continuous medical images; storing at least one of location information, features and diagnosis of the diagnosed lesion; and generating the feedback information to output the diagnosis of the diagnosed lesion. ### Claim 27 (depends on 26) The lesion diagnosis method of claim 26, wherein the generating of the feedback information further comprises, based on at least one of the location information and the features, generating visual information to notify that the lesion has been diagnosed. ### Claim 28 (depends on 27) The lesion diagnosis method of claim 27, wherein the diagnosing of a lesion comprises, based on at least one of the location information and the features, an undiagnosed a lesion.