WO2019200740A1 - 肺结节的检测方法、装置、计算机设备和存储介质 - Google Patents

肺结节的检测方法、装置、计算机设备和存储介质 Download PDF

Info

Publication number
WO2019200740A1
WO2019200740A1 PCT/CN2018/095459 CN2018095459W WO2019200740A1 WO 2019200740 A1 WO2019200740 A1 WO 2019200740A1 CN 2018095459 W CN2018095459 W CN 2018095459W WO 2019200740 A1 WO2019200740 A1 WO 2019200740A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
lung
dimensional
nodule
nodules
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2018/095459
Other languages
English (en)
French (fr)
Inventor
刘新卉
刘莉红
吴天博
马进
王健宗
肖京
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2019200740A1 publication Critical patent/WO2019200740A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three-dimensional [3D] modelling for computer graphics
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30061—Lung
    • G06T2207/30064—Lung nodule

Definitions

  • the present application relates to the field of computer technology, and in particular, to a method, device, computer device and storage medium for detecting lung nodules.
  • Sarcoidosis is a multi-system, multi-organ granulomatous disease of unknown etiology that has recently attracted widespread attention in the country. Pulmonary sarcoidosis often invades the lungs, bilateral hilar lymph nodes, eyes, skin and other organs. Pulmonary cell proliferation or foreign bodies can lead to the production of pulmonary nodules. In an increasingly deteriorating environment, more and more people have lung nodules in their lungs.
  • pulmonary nodules are already a very common symptom, and many young people need to go to the hospital to remove lung nodules as soon as possible.
  • the main purpose of the present application is to provide a method, a device, a computer device and a storage medium for detecting lung nodules, which are suitable for segmenting lung regions of all lung CT images and overcome the slow detection speed in the prior art. defect.
  • the present application provides a method for detecting a pulmonary nodule, comprising the following steps:
  • the lung CT image is segmented by a three-dimensional convolutional neural network segmentation model to segment the lung region image;
  • Suspicious nodules are detected from the image of the lung region by a three-dimensional U-Net detection model
  • the suspected nodules are classified by a three-dimensional two-class network to remove false nodules.
  • the application also provides a detecting device for a pulmonary nodule, comprising:
  • a segmentation unit for segmenting a lung CT image by a three-dimensional convolutional neural network segmentation model, and segmenting the lung region image
  • a detecting unit configured to detect a suspicious nodule from the image of the lung region by using a three-dimensional U-Net detection model
  • a classification unit for classifying the suspected nodules by a three-dimensional two-class network to remove false nodules.
  • the application further provides a computer device comprising a memory and a processor, the memory storing computer readable instructions, the processor executing the computer readable instructions to implement the steps of any of the methods described above.
  • the present application also provides a computer non-transitory readable storage medium having stored thereon computer readable instructions that, when executed by a processor, implement the steps of any of the methods described above.
  • the method, device, computer equipment and storage medium for detecting pulmonary nodules provided in the present application divide the lung CT image by a three-dimensional convolutional neural network segmentation model, and segment the lung region image, and the segmentation speed is fast, so that the subsequent The speed of detecting pulmonary nodules is accelerated; at the same time, it can be applied to segmentation of the lung region of all lung CT images; suspicious nodules are detected from the lung region images by a three-dimensional U-Net detection model, and The suspicious nodules are classified by a three-dimensional two-class network, the pseudo-nodules are removed, the effect of detecting lung nodules is improved, and the detection accuracy is improved.
  • FIG. 1 is a schematic view showing the steps of a method for detecting a pulmonary nodule according to an embodiment of the present application
  • FIG. 2 is a schematic view showing the steps of a method for detecting a pulmonary nodule in another embodiment of the present application
  • FIG. 3 is a schematic structural view of a lung nodule detecting device according to an embodiment of the present application.
  • FIG. 4 is a block diagram showing the structure of a detecting unit in an embodiment of the present application.
  • FIG. 5 is a structural block diagram of a classification unit in another embodiment of the present application.
  • FIG. 6 is a schematic block diagram showing the structure of a computer device according to an embodiment of the present application.
  • a method for detecting a pulmonary nodule is provided in the embodiment of the present application, which includes the following steps:
  • Step S1 segmenting the lung CT image by a three-dimensional convolutional neural network segmentation model, and segmenting the lung region image;
  • Step S2 detecting suspicious nodules from the image of the lung region through a three-dimensional U-Net detection model
  • step S3 the suspicious nodules are classified by a three-dimensional two-class network to remove false nodules.
  • the hospital uses a medical device to take a CT image of the patient's lungs, the lung CT image being a three-dimensional image.
  • the CT image of the lung includes not only the image of the lung region, but also some other tissues around the lung or images of medical equipment, which may interfere with the detection process of the subsequent pulmonary nodules, and it is also easy to increase the detection of pulmonary nodules.
  • the search area When the search area.
  • step S1 when detecting pulmonary nodules from the patient's lung CT image, in order to reduce interference of other tissues and unnecessary search areas, it is usually necessary to first segment the lung CT image and divide it from An image of the lung area.
  • an image segmentation algorithm for example, threshold segmentation, region segmentation, edge segmentation, histogram method, etc.
  • the image segmentation algorithm not only has a slow segmentation speed, but also cannot Suitable for segmentation of all lung CT images.
  • the segmentation of the image of the lung region can also be achieved by the two-dimensional segmentation model, but the segmentation effect is not ideal. Therefore, in the step S1 of the embodiment, the segmentation model of the three-dimensional full convolutional neural network is used to segment the lung CT image to realize three-dimensional segmentation, and the lung region image is segmented from the lung CT image to reduce other tissues. Interference and unnecessary search areas.
  • the difference between the three-dimensional convolutional neural network segmentation model and the two-dimensional segmentation model is that the two-dimensional segmentation model divides the three-dimensional lung CT image into two-dimensional images, which inevitably results in incomplete data expression and affects the segmentation effect. Dimensional segmentation needs to be segmented from multiple dimensions, resulting in slower segmentation speed.
  • the three-dimensional convolutional neural network segmentation model is used to directly realize three-dimensional segmentation.
  • the segmented lung region image is still a three-dimensional image.
  • the data is comprehensive and the segmentation effect is good. , the segmentation speed is fast.
  • the segmentation of the lung region image using the three-dimensional convolutional neural network segmentation model has two major advantages over the segmentation using the existing image segmentation algorithm:
  • the segmentation speed is faster. It takes 2-8 minutes for the traditional image algorithm to segment each lung CT image, and it takes only 5-10 seconds to segment the model using the 3D convolutional neural network.
  • the traditional image segmentation algorithm cannot segment all lung CT images into the lung region (for example, some special lung CT images cannot be segmented), which seriously hinders the subsequent detection process and uses three-dimensional convolutional nerves.
  • the network segmentation model ensures that all lung CT images are segmented.
  • the lung region image is segmented from the lung CT image; as described in step S2, the suspicious nodule is detected from the lung region image by the three-dimensional U-Net detection model, which is suspicious Nodules refer to features that are suspected to be pulmonary nodules.
  • Two-dimensional U-Net is an existing segmentation model for two-dimensional images. It is a semantic segmentation network based on FCN. It is suitable for segmentation of medical images. It can also segment three-dimensional images, but the segmentation effect. Not good, therefore, a three-dimensional U-Net detection model is used in this embodiment for detecting suspicious nodules.
  • the suspicious nodules detected in the above step S2 include many false nodules (false positives). If these false nodules are not removed, a lot of unnecessary work will be brought to the doctor, and therefore, in order to ensure a high lung knot. For the detection rate, the suspicious nodules need to be further screened.
  • step S3 of this embodiment the suspicious nodules are further classified using a three-dimensional two-class network, and the confidence of each suspicious nodule is obtained, and the lower confidence is removed as a false nodule (ie, the pseudo-nodule is removed) Section), only the higher confidence is reserved as a true nodule, so as to achieve the purpose of suppressing false positives and improve the detection rate of pulmonary nodules.
  • step S3 other two-dimensional classification models may also be used for classification, but the classification effect is not good, and the classification effect is better by using the three-dimensional two-class network.
  • the step of segmenting the lung CT image by the three-dimensional convolutional neural network segmentation model and segmenting the lung region image before the step S1 includes:
  • Step S101 preprocessing the lung CT image to remove image noise.
  • the pre-processing process is a conventional method.
  • the pre-processing process may be: binarizing the CT image of the lung using -600 HU as a threshold, because the upper and lower regions of the CT image of the lung are generally Some slices are connected to the outside world and need to be removed; therefore, the final image pixel values are cropped to [-1200,600] and then scaled to [0,255]; the non-lung area pixels are set to 170.
  • the pre-processing described above may also adjust the pixel spacing, image contrast, etc. of the lung CT image.
  • the step S1 of segmenting the lung CT image by the three-dimensional convolutional neural network segmentation model includes:
  • the lung region features in the lung CT image are extracted using multiple sets of convolutional layers, and a batch normalization method is added to each set of convolutional layers to convolve each set of convolutional layers and to upsample the convolution results. Processing to obtain an image of the lung region consistent with the original size of the lung CT image.
  • the segmentation model of the above three-dimensional convolutional neural network has a structure of a plurality of convolution layers similar to a VGG network (a convolutional neural network), and a batch is added.
  • the normalization layer convolves and upsamples each convolutional layer.
  • upsampling is the inverse process of downsampling.
  • Upsampling and downsampling are all re-collecting the digital signal.
  • the sampling rate of the re-acquisition is compared with the sampling rate of the original obtained digital signal (such as sampling from the analog signal).
  • the larger than the original signal is called upsampling, and the smaller is called downsampling.
  • the essence of upsampling is interpolation or interpolation.
  • the lung region features in the lung CT image are extracted using 5 sets of convolution layers, and a 1X1X1 convolution is performed on each convolutional layer by adding a batch normalization method to each set of convolutional layers. And after performing upsampling processing on the convolution result, obtaining an image of the lung region that is consistent with the original size of the lung CT image.
  • the loss function used in the three-dimensional U-Net detection model in the above step S2 is a focal loss function and a regression loss function.
  • Use focal The purpose of the loss function is to solve the problem of serious imbalance between positive and negative samples during the training of the test model.
  • the simple negative sample and the negative negative sample have the same contribution to the model.
  • the focal loss can make the model focus on the negative sample differentiation of the difficulty, so that the model result has better performance on the FROC curve.
  • the step S2 of detecting a suspicious nodule from the image of the lung region by the three-dimensional U-Net detection model includes:
  • the image of the lung region is sequentially subjected to four convolution and max pooling (convolution and pooling), and two deconvolution (reconvolution) calculations to obtain a first probability map; and the image in the lung region passes Two branches are added before the deconvolution calculation, and the two branches are respectively calculated by the corresponding deconvolution to obtain a corresponding second probability map;
  • the two second probability maps and the first probability map are simultaneously input into the back propagation algorithm for iterative calculation to obtain a final probability map, where the final probability map represents the lung nodules in the lung region image.
  • a three-dimensional U-Net detection model is used, which is limited to the GPU (graphics processor) storage size, and needs to perform sample processing on the segmented lung region image. Specifically, when inputting an image, select input 128*128. *128 cube, at this time, 70% of the input cube has a nodule, and the remaining 30% take random cropping, so that the sample contains the background sample. Large lung nodules are less numerous than small pulmonary nodules, so when the sample is removed, the nodule samples larger than 30 mm and 40 mm in diameter are expanded 2-6 times. At the same time, all of the above samples need to be scaled by probability to eliminate overfitting problems.
  • the 128*128*128 cube image subjected to the sample processing described above is input into the three-dimensional U-Net detection model, and after four calculations of convolution and max pooling, 2 deconvolution calculations get the first probability map of 32*32*32; at the same time, add new branches before each deconvolution calculation, and each branch calculates the second probability map of output 32*32*32 through the corresponding deconvolution layer;
  • two 32*32*32 second probability maps and 32*32*32 first probability maps are simultaneously input into the back propagation algorithm for iterative calculation to obtain a final probability map,
  • the final probability map represents the probability of a pulmonary nodule in the image of the lung region; according to the final probability map, a suspicious nodule is detected from the image of the lung region; if the probability exceeds a preset value, it is regarded as a suspicious knot Section.
  • an IOU Intersection-over-Union index
  • the IOU is an evaluation index for the segmentation model. Is the ratio of the intersection and the union of the detection area and the real area. The larger the value obtained, the closer the segmentation result of the segmentation model is to the real value.
  • the foreground here refers to the target, that is, the pulmonary nodules, the IOU is greater than 0.5 for the foreground, and the less than 0.02 is the background.
  • the step S3 of classifying the suspected nodules by using the three-dimensional two-class network to remove the pseudo-nodules includes:
  • the feature map is sequentially input to three fully connected layers, and is input to the softmax layer via the fully connected layer, and the probability of the suspicious nodule classification is obtained through the softmax layer output, and the suspicious nodule is classified. Probability is the confidence that a suspected nodule is detected as a pulmonary nodule;
  • the pseudo nodule is removed, and the pseudo nodule is a suspicious nodule with a confidence lower than a set value.
  • the step of sequentially inputting the feature map to three fully connected layers includes:
  • the network structure of the three-dimensional two-class network includes five convolution layers, three fully-connected layers, and one batch. Normalization layer and softmax layer.
  • the suspected nodule detected in step S2 is input into a three-dimensional two-class network in a cube of 36 ⁇ 36 ⁇ 20 size, and a feature map is obtained through five convolution layers, as described in S302.
  • the feature map is input to the batch normalization layer, it is sequentially input to the three fully connected layers, and finally the probability of the suspicious nodule classification is obtained through the softmax output, and the probability of the suspicious nodule classification is detected as a suspicious nodule as a pulmonary nodule.
  • Confidence It is preset to have a set value.
  • the confidence level When the confidence level is higher than the set value, it is judged to be a pulmonary nodule; if the confidence level is lower than the above set value, it is judged to be a false nodule, that is, a false positive, which needs to be removed. In order to achieve the purpose of suppressing false positives, thereby ensuring the detection rate of lung nodules in the present embodiment and reducing the workload of doctors.
  • the lung nodule position can be determined, and the three-dimensional shape of the pulmonary nodule is depicted in the three-dimensional image, and the calculation is performed.
  • the size of the pulmonary nodules is sent to the medical computer device for display, so that the doctor can make a reasonable treatment according to the above information. Due to the different information such as the size, shape and location of the lung nodules, the doctor's treatment plan will be different. Therefore, obtaining the above information will help the doctor to make a reasonable treatment plan.
  • the above-mentioned symptoms of the patient and the corresponding treatment plan can also be recorded in a database for preservation.
  • the CT image of the lung is segmented by a three-dimensional convolutional neural network segmentation model, and the lung region image is segmented, and the segmentation speed is fast, so that the subsequent The speed of detecting pulmonary nodules is accelerated; at the same time, it can be applied to segmentation of the lung region of all lung CT images; suspicious nodules are detected from the lung region images by a three-dimensional U-Net detection model, and The suspicious nodules are classified by a three-dimensional two-class network, the pseudo-nodules are removed, the effect of detecting lung nodules is improved, and the detection accuracy is improved.
  • an embodiment of the present application further provides a device for detecting a pulmonary nodule, including:
  • the dividing unit 10 is configured to segment the lung CT image by using a three-dimensional convolutional neural network segmentation model, and segment the lung region image;
  • the detecting unit 20 is configured to detect a suspicious nodule from the image of the lung region by using a three-dimensional U-Net detection model
  • the classification unit 30 is configured to classify the suspected nodules by a three-dimensional two-class network to remove false nodules.
  • the hospital uses a medical device to take a CT image of the patient's lungs, the lung CT image being a three-dimensional image.
  • the CT image of the lung includes not only the image of the lung region, but also some other tissues around the lung or images of medical equipment, which may interfere with the detection process of the subsequent pulmonary nodules, and it is also easy to increase the detection of pulmonary nodules.
  • the search area When the search area. Therefore, when detecting pulmonary nodules from the patient's lung CT image, in order to reduce interference of other tissues and unnecessary search areas, the segmentation unit 10 usually needs to segment the lung CT image first, and segment the lung region therefrom. image.
  • an image segmentation algorithm for example, threshold segmentation, region segmentation, edge segmentation, histogram method, etc.
  • the image segmentation algorithm not only has a slow segmentation speed, but also cannot Suitable for segmentation of all lung CT images.
  • the segmentation of the image of the lung region can also be achieved by the two-dimensional segmentation model, but the segmentation effect is not ideal. Therefore, in the segmentation unit 10 of the present embodiment, a segmentation model of a three-dimensional full convolutional neural network is used to segment a lung CT image to realize three-dimensional segmentation, and a lung region image is segmented from a lung CT image to reduce other tissues. Interference and unnecessary search areas.
  • the difference between the three-dimensional convolutional neural network segmentation model and the two-dimensional segmentation model is that the two-dimensional segmentation model divides the three-dimensional lung CT image into two-dimensional images, which inevitably results in incomplete data expression and affects the segmentation effect. Dimensional segmentation needs to be segmented from multiple dimensions, resulting in slower segmentation speed.
  • the three-dimensional convolutional neural network segmentation model is used to directly realize three-dimensional segmentation.
  • the segmented lung region image is still a three-dimensional image. The data is comprehensive and the segmentation effect is good. , the segmentation
  • the segmentation of the lung region image using the three-dimensional convolutional neural network segmentation model in the segmentation unit 10 has two major advantages over the segmentation using the existing image segmentation algorithm:
  • the segmentation speed is faster. It takes 2-8 minutes for the traditional image algorithm to segment each lung CT image, and it takes only 5-10 seconds to segment the model using the 3D convolutional neural network.
  • the traditional image segmentation algorithm cannot segment all lung CT images into the lung region (for example, some special lung CT images cannot be segmented), which seriously hinders the subsequent detection process and uses three-dimensional convolutional nerves.
  • the network segmentation model ensures that all lung CT images are segmented.
  • the lung region image is segmented from the lung CT image; the detecting unit 20 detects the suspicious nodule from the lung region image by the three-dimensional U-Net detection model, which is suspicious Nodules refer to features that are suspected to be pulmonary nodules.
  • U-Net is an existing segmentation model for 2D images. There are currently other 2D detection models, such as faster-rcnn, ssd, etc., which can also be used to detect suspicious nodules;
  • the image is three-dimensional data, and it is apparent that the use of the three-dimensional detection model to detect suspicious nodules is the best; therefore, the three-dimensional U-Net detection model is used in this embodiment to detect suspicious nodules.
  • the use of the three-dimensional U-Net detection model for suspicious nodules in this application is intended to be more effective for other detection models.
  • the suspicious nodules detected by the detecting unit 20 include many false nodules (false positives). If these false nodules are not removed, a lot of unnecessary work will be brought to the doctor, and therefore, in order to ensure a high lung knot. For the detection rate, the suspicious nodules need to be further screened.
  • the classification unit 30 of the present embodiment further classifies the suspected nodules using a three-dimensional two-class network to obtain the confidence of each suspicious nodule, and removes the false nodule as a false nodule (ie, removes the pseudo-nodule) ), only the higher confidence is retained as a true nodule, so as to achieve the purpose of suppressing false positives and improve the detection rate of pulmonary nodules.
  • a three-dimensional two-class network to obtain the confidence of each suspicious nodule, and removes the false nodule as a false nodule (ie, removes the pseudo-nodule) ), only the higher confidence is retained as a true nodule, so as to achieve the purpose of suppressing false positives and improve the detection rate of pulmonary nodules.
  • step S3 other two-dimensional classification models may also be used for classification, but the classification effect is not good, and the classification effect is better by using the three-dimensional two-class network.
  • the detecting device for the pulmonary nodule further includes:
  • the pre-processing unit 101 is configured to pre-process the lung CT image to remove image noise.
  • the pre-processing process is a conventional method.
  • the pre-processing process may be: binarizing the CT image of the lung using -600 HU as a threshold, because the upper and lower regions of the CT image of the lung are generally Some slices are connected to the outside world and need to be removed; therefore, the final image pixel values are cropped to [-1200,600] and then scaled to [0,255]; the non-lung area pixels are set to 170.
  • the pixel interval, image contrast, and the like of the lung CT image may be adjusted during the pre-processing of the pre-processing unit 101.
  • the dividing unit 10 is specifically configured to:
  • the lung region features in the lung CT image are extracted using multiple sets of convolutional layers, and a batch normalization method is added to each set of convolutional layers to convolve each set of convolutional layers and to upsample the convolution results. Processing to obtain an image of the lung region consistent with the original size of the lung CT image.
  • the segmentation model of the three-dimensional convolutional neural network has a structure of a plurality of convolution layers similar to the VGG network, and a batch normalization layer is added to each volume of the convolutional layer.
  • Product sum upsampling is the inverse process of downsampling.
  • Upsampling and downsampling are all re-collecting the digital signal. The sampling rate of the re-acquisition is compared with the sampling rate of the original obtained digital signal (such as sampling from the analog signal). The larger than the original signal is called upsampling, and the smaller is called downsampling.
  • the essence of upsampling is interpolation or interpolation.
  • the lung region features in the lung CT image are extracted using 5 sets of convolution layers, and a 1X1X1 convolution is performed on each convolutional layer by adding a batch normalization method to each set of convolutional layers. And after performing upsampling processing on the convolution result, obtaining an image of the lung region that is consistent with the original size of the lung CT image.
  • the loss function used by the three-dimensional U-Net detection model in the detection unit 20 is a focal loss function and a regression loss function.
  • Use focal The purpose of the loss function is to solve the problem of serious imbalance between positive and negative samples during the training of the test model.
  • the simple negative sample and the negative negative sample have the same contribution to the model.
  • the focal loss can make the model focus on the negative sample differentiation of the difficulty, so that the model result has better performance on the FROC curve.
  • the detecting unit 20 includes:
  • a first calculating module 201 configured to perform a first probability map on the lung region image by four times convolution and max pooling, and two deconvolution calculations; and before the image of the lung region is subjected to two deconvolution calculations respectively Adding a branch, and the two branches are respectively calculated by corresponding deconvolution to obtain a corresponding second probability map;
  • a second calculation module 202 configured to simultaneously input the two second probability maps and the first probability map into the back propagation algorithm for iterative calculation to obtain a final probability map, where the final probability map represents the lungs The probability of a pulmonary nodule in the regional image;
  • the detecting module 203 is configured to detect a suspicious nodule from the lung region image according to the final probability map.
  • a three-dimensional U-Net detection model is used, which is limited to the GPU storage size, and needs to perform sample processing on the segmented lung region image. Specifically, when inputting an image, a stereotype of 128*128*128 is selected. At this point, 70% of the input cubes have a nodule, and the remaining 30% are randomly cropped to include a background sample in the sample. Large lung nodules are less numerous than small pulmonary nodules, so when the sample is removed, the nodule samples larger than 30 mm and 40 mm in diameter are expanded 2-6 times. At the same time, all of the above samples need to be scaled by probability to eliminate overfitting problems.
  • the first calculation module 201 inputs the 128*128*128 cube image subjected to the sample processing described above into the three-dimensional U-Net detection model, and after two times of convolution and max pooling calculation, two deconvolutions are performed.
  • the module 202 simultaneously inputs two 32*32*32 second probability maps and 32*32*32 first probability maps into the back propagation algorithm for iterative calculation to obtain a final probability map, and the final probability map represents The probability of a pulmonary nodule in the lung region image; a suspicious nodule is detected from the lung region image based on the final probability map.
  • the detecting module 203 needs to identify the foreground and the background.
  • the IOU index is used to determine the foreground and the background.
  • the IOU is an evaluation index for the segmentation model, and is the detection region and the real region. The ratio of intersection to union, the larger the value obtained, the closer the segmentation result of the segmentation model is to the real value.
  • the foreground here refers to the target, that is, the pulmonary nodules, the IOU is greater than 0.5 for the foreground, and the less than 0.02 is the background.
  • the classification unit 30 includes:
  • a convolution module 301 configured to input the suspicious nodule to an input layer of a three-dimensional two-class network, and sequentially obtain a feature map through five convolution layers;
  • the output module 302 is configured to sequentially input the feature map to three fully connected layers, input to the softmax layer via the fully connected layer, and finally obtain the probability of the suspicious nodule classification through the softmax layer output.
  • the probability of classifying suspicious nodules is the confidence that the suspected nodule is detected as a pulmonary nodule;
  • the removing module 303 is configured to remove the pseudo nodule, which is a suspicious nodule with a confidence lower than a set value.
  • the outputting the module 302 to sequentially input the feature map to the three fully connected layers specifically includes:
  • the feature map is input to the batch normalization layer, and sequentially input to the three fully connected layers via the batch normalization layer.
  • the network structure of the three-dimensional two-class network includes five convolution layers, three full-connection layers, and a batch normalization layer and a softmax layer.
  • the convolution module 301 inputs the detected suspicious nodules into a three-dimensional two-class network in a cube of 36 ⁇ 36 ⁇ 20 size, obtains a feature map through five convolution layers, and the output module 302 inputs the feature map to batch normalization. After the layer, it is sequentially input to the three fully connected layers, and finally the probability of the suspicious nodule classification is obtained through the softmax output, and the probability of the suspected nodule classification is detected as the suspicious nodule as the confidence of the pulmonary nodule. It is preset to have a set value.
  • the removal module 303 removes it to achieve the purpose of suppressing false positives, thereby ensuring the detection rate of the pulmonary nodules in the present embodiment and reducing the workload of the doctor.
  • the lung nodule position can be determined, and the three-dimensional shape of the pulmonary nodule is depicted in the three-dimensional image, and the calculation is performed.
  • the size of the pulmonary nodules is sent to the medical computer device for display, so that the doctor can make a reasonable treatment according to the above information. Due to the different information such as the size, shape and location of the lung nodules, the doctor's treatment plan will be different. Therefore, obtaining the above information will help the doctor to make a reasonable treatment plan.
  • the above-mentioned symptoms of the patient and the corresponding treatment plan can also be recorded in a database for preservation.
  • the dividing unit 10 divides the CT image of the lung by a three-dimensional convolutional neural network segmentation model, and segments the lung region image, and the segmentation speed is fast.
  • detection unit 20 detects from the lung region images through a three-dimensional U-Net detection model Suspicious nodules are generated, and the classification unit 30 classifies the suspected nodules through a three-dimensional two-class network, removes false nodules, improves the effect of detecting lung nodules, and improves detection accuracy.
  • the computer device may be a server, and its internal structure may be as shown in FIG. 6.
  • the computer device includes a processor, memory, network interface, and database connected by a system bus. Among them, the computer designed processor is used to provide calculation and control capabilities.
  • the memory of the computer device includes a non-volatile storage medium, an internal memory.
  • the non-volatile storage medium stores an operating system, computer readable instructions, and a database.
  • the internal memory provides an environment for operation of an operating system and computer readable instructions in a non-volatile storage medium.
  • the database of the computer device is used to store data such as a three-dimensional convolutional neural network segmentation model.
  • the network interface of the computer device is used to communicate with an external terminal via a network connection.
  • the computer readable instructions are executed by a processor to implement a method of detecting a pulmonary nodule.
  • the processor performs the steps of the method for detecting a lung nodule described above: segmenting a lung CT image by a three-dimensional convolutional neural network segmentation model, segmenting a lung region image; and extracting the lung from the lung through a three-dimensional U-Net detection model Suspicious nodules are detected in the partial region image; the suspected nodules are classified by a three-dimensional two-class network to remove the pseudo-nodules.
  • the processor segments the lung CT image by using a three-dimensional convolutional neural network segmentation model, and the step of segmenting the lung region image includes:
  • the lung CT image is pre-processed to remove image noise.
  • the processor segments the lung CT image by using a three-dimensional convolutional neural network segmentation model, including:
  • the lung region features in the lung CT image are extracted using multiple sets of convolutional layers, and a batch normalization method is added to each set of convolutional layers to convolve each set of convolutional layers and to upsample the convolution results. Processing to obtain an image of the lung region consistent with the original size of the lung CT image.
  • the loss function used by the three-dimensional U-Net detection model is a focal loss function and a regression loss function.
  • the processor detects a suspicious nodule from the image of the lung region through a three-dimensional U-Net detection model, including:
  • the lung region image is sequentially subjected to four convolution and max pooling, and two deconvolution calculations to obtain a first probability map; and a branch is added to the lung region image before undergoing two deconvolution calculations, respectively.
  • the branches are respectively calculated by the corresponding deconvolution to obtain a corresponding second probability map;
  • Suspicious nodules are detected from the lung region image based on the final probability map.
  • the processor classifies the suspected nodules through a three-dimensional two-class network, and the steps of removing the pseudo-nodules include:
  • the feature map is sequentially input to three fully connected layers, and then input to the softmax layer via the fully connected layer, and finally the probability of the suspicious nodule classification is obtained through the softmax layer output, the suspicious nodule classification Probability is the confidence that a suspected nodule is detected as a pulmonary nodule;
  • the pseudo nodule is removed, which is a suspicious nodule with a confidence lower than the set value.
  • the step of the processor sequentially inputting the feature map to three fully connected layers includes:
  • the feature map is input to the batch normalization layer, and sequentially input to the three fully connected layers via the batch normalization layer.
  • FIG. 6 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation of the computer device to which the solution of the present application is applied.
  • An embodiment of the present application further provides a computer non-volatile readable storage medium having stored thereon computer readable instructions, which are implemented by a processor to implement a method for detecting a pulmonary nodule, specifically:
  • the lung CT image is segmented by a three-dimensional convolutional neural network segmentation model to segment the lung region image; suspicious nodules are detected from the lung region image through a three-dimensional U-Net detection model;
  • the classification network classifies the suspected nodules to remove false nodules.
  • the processor segments the lung CT image by using a three-dimensional convolutional neural network segmentation model, and the step of segmenting the lung region image includes:
  • the lung CT image is pre-processed to remove image noise.
  • the processor segments the lung CT image by using a three-dimensional convolutional neural network segmentation model, including:
  • the lung region features in the lung CT image are extracted using multiple sets of convolutional layers, and a batch normalization method is added to each set of convolutional layers to convolve each set of convolutional layers and to upsample the convolution results. Processing to obtain an image of the lung region consistent with the original size of the lung CT image.
  • the loss function used by the three-dimensional U-Net detection model is a focal loss function and a regression loss function.
  • the processor detects a suspicious nodule from the image of the lung region through a three-dimensional U-Net detection model, including:
  • the lung region image is sequentially subjected to four convolution and max pooling, and two deconvolution calculations to obtain a first probability map; and a branch is added to the lung region image before undergoing two deconvolution calculations, respectively.
  • the branches are respectively calculated by the corresponding deconvolution to obtain a corresponding second probability map;
  • Suspicious nodules are detected from the lung region image based on the final probability map.
  • the processor classifies the suspected nodules through a three-dimensional two-class network, and the steps of removing the pseudo-nodules include:
  • the feature map is sequentially input to three fully connected layers, and then input to the softmax layer via the fully connected layer, and finally the probability of the suspicious nodule classification is obtained through the softmax layer output, the suspicious nodule classification Probability is the confidence that a suspected nodule is detected as a pulmonary nodule;
  • the pseudo nodule is removed, which is a suspicious nodule with a confidence lower than the set value.
  • the step of the processor sequentially inputting the feature map to three fully connected layers includes:
  • the feature map is input to the batch normalization layer, and sequentially input to the three fully connected layers via the batch normalization layer.
  • the method, device, computer equipment and storage medium for detecting pulmonary nodules divide the lung CT image by a three-dimensional convolutional neural network segmentation model, and segment the lung region.
  • the image, the segmentation speed is fast, so that the speed of subsequent detection of the pulmonary nodules is accelerated; at the same time, it can be applied to segmentation of the lung region of all lung CT images; from the lung region image through the three-dimensional U-Net detection model Suspicious nodules are detected, and the suspected nodules are classified by a three-dimensional two-class network, the pseudo-nodules are removed, the effect of detecting lung nodules is improved, and the detection accuracy is improved.
  • Non-volatile memory can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory can include random access memory (RAM) or external cache memory.
  • RAM is available in a variety of formats, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronization.
  • SRAM static RAM
  • DRAM dynamic RAM
  • SDRAM synchronous DRAM
  • SSRSDRAM dual-speed SDRAM
  • ESDRAM enhanced SDRAM
  • SLDRAM Link (Synchlink) DRAM
  • SLDRAM Memory Bus
  • RDRAM Direct RAM
  • DRAM Direct Memory Bus Dynamic RAM
  • RDRAM Memory Bus Dynamic RAM

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Graphics (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Geometry (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

一种肺结节的检测方法、装置、计算机设备和存储介质，所述方法包括：通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像（S1）；通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节（S2）；通过三维的二分类网络对所述可疑结节进行分类，去除假结节（S3）。所述方法适用于对所有肺部CT图像进行肺部区域的分割。

Description

肺结节的检测方法、装置、计算机设备和存储介质
本申请要求于2018年4月20日提交中国专利局、申请号为2018103621990，发明名称为“肺结节的检测方法、装置、计算机设备和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机技术领域，特别涉及一种肺结节的检测方法、装置、计算机设备和存储介质。
 背景技术
肺结节病(sarcoidosis)是一种病因未明的多系统多器官的肉芽肿性疾病，近来已引起国内广泛注意。肺结节病常侵犯肺、双侧肺门淋巴结、眼、皮肤等器官。肺部细胞增生或异物都会导致肺结节的产生，在日益变差的环境中，越来越多的人肺部产生了肺结节。现如今，肺结节已经是非常常见的一种症状，很多年轻人都需要去医院尽早摘除肺结节。
肺结节摘除之前需要从肺部CT图像中分离出肺结节部分，目前通常是依靠医生的临床经验进行分离，或者一些现有的图像分割算法进行分离，其诊断速度较慢，需要花费医生较多时间，而且准确率较低；同时，现有的检测方法不能对一些较为特殊的肺部CT分割出肺部区域，严重阻碍了后续的检测。
 技术问题
本申请的主要目的为提供一种肺结节的检测方法、装置、计算机设备和存储介质，适用于对所有肺部CT图像进行肺部区域的分割，并克服了现有技术中检测速度慢的缺陷。
 技术解决方案
为实现上述目的，本申请提供了一种肺结节的检测方法，包括以下步骤：
通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；
通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；
通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
本申请还提供了一种肺结节的检测装置，包括：
分割单元，用于通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；
检测单元，用于通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；
分类单元，用于通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
本申请还提供一种计算机设备，包括存储器和处理器，所述存储器存储有计算机可读指令，所述处理器执行所述计算机可读指令时实现上述任一项所述方法的步骤。
本申请还提供一种计算机非易失性可读存储介质，其上存储有计算机可读指令，所述计算机可读指令被处理器执行时实现上述任一项所述的方法的步骤。
 有益效果
本申请中提供的肺结节的检测方法、装置、计算机设备和存储介质，通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像，分割速度快，使得后续检测出肺结节的速度加快；同时，可适用于对所有肺部CT图像进行肺部区域的分割；通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节，以及通过三维的二分类网络对所述可疑结节进行分类，去除假结节，提高检测肺结节的效果，以及提升检测准确率。
 附图说明
图1 是本申请一实施例中肺结节的检测方法步骤示意图；
图2 是本申请另一实施例中肺结节的检测方法步骤示意图；
图3 是本申请一实施例中肺结节的检测装置结构示意图；
图4 是本申请一实施例中的检测单元结构框图；
图5 是本申请另一实施例中的分类单元结构框图；
图6 为本申请一实施例的计算机设备的结构示意框图。
 本发明的最佳实施方式
参照图1，本申请实施例中提供了一种肺结节的检测方法，包括以下步骤：
步骤S1，通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；
步骤S2，通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；
步骤S3，通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
在本实施例中，医院使用医疗设备拍摄病人肺部的CT图像，该肺部CT图像为三维图像。肺部CT图像中不仅包括肺部区域图像，还包括肺部周围的一些其它组织或者医疗设备图像等，其都可能对后续肺结节的检测过程造成干扰，同时也容易增大检测肺结节时的搜索区域。
因此，如上述步骤S1所述，在从病人的肺部CT图像中检测肺结节时，为了减少其它组织的干扰和不必要的搜索区域，通常需要先对肺部CT图像进行分割，从中分割出肺部区域图像。现有技术中通常使用图像分割算法（例如阈值分割、区域分割、边缘分割、直方图法等）从肺部CT图像中分割出肺部区域图像，但是，图像分割算法不仅分割速度慢，而且无法适用于对所有的肺部CT图像进行分割。
通过二维的分割模型也可以实现肺部区域图像的分割，但是分割效果不理想。因此，本实施例的步骤S1中使用了三维的全卷积神经网络的分割模型对肺部CT图像进行分割来实现三维分割，从肺部CT图像中分割出肺部区域图像，减少其它组织的干扰和不必要的搜索区域。三维的卷积神经网络分割模型相对于二维的分割模型，其区别在于，二维分割模型将三维的肺部CT图像分割我二维图像，必然造成数据表达不全面，影响分割效果，且二维分割时需要从多个维度进行分割，造成分割速度较慢；而使用三维的卷积神经网络分割模型直接实现三维分割，分割出来的肺部区域图像还是三维图像，数据表达全面，分割效果好，分割速度快。
因此，在本步骤的S1中，使用三维的卷积神经网络分割模型进行肺部区域图像的分割，相比于使用现有图像分割算法来分割具有两大优势：
①、分割速度更快，传统图像算法对每一个肺部CT图像进行分割需要使用2-8分钟的时间，而使用三维的卷积神经网络分割模型只需要5-10秒。
②、传统图像分割算法不能对所有的肺部CT图像分割出肺部区域（例如一些较为特殊的肺部CT图像则无法进行分割），严重阻碍了后续的检测过程，而使用三维的卷积神经网络分割模型则能保证对所有肺部CT图像进行分割处理。
在经过上述步骤S1之后，则从肺部CT图像中分割出肺部区域图像；如步骤S2中所述则通过三维的U-Net检测模型从上述肺部区域图像中检测出可疑结节，可疑结节指的是疑似为肺结节的特征。二维U-Net是一种现有的针对二维图像的分割模型，其为基于FCN的一个语义分割网络，适合用来做医学图像的分割，其也可以对三维图像进行分割，只是分割效果不好，因此，本实施例中使用了三维的U-Net检测模型用于检测可疑结节。当前还有一些其它二维的检测模型，比如faster-rcnn，ssd等也可以用于检测出可疑结节；由于上述肺部区域图像是三维数据，显然，使用三维的检测模型检测可疑结节的效果最好；因此，本实施例中使用三维的U-Net检测模型从中检测可疑结节。经过试验对比，本申请中使用三维的U-Net检测模型进行可疑结节的检测想对于其它检测模型的效果更好。
上述步骤S2中检测出的可疑结节中包括有许多假结节（假阳性），若不去除这些假结节，将对医生带来大量不必要的工作，因此，为了保证较高的肺结节检出率，则需要对可疑结节进行进一步地筛选。
本实施例的步骤S3中，使用三维的二分类网络对所述可疑结节进行进一步地分类，得到每个可疑结节的置信度，去掉置信度较低的作为假结节（即去除假结节），只保留置信度较高的作为真结节，以此达到抑制假阳性的目的，提高肺结节检出率。在步骤S3中，也可以使用其它二维的分类模型进行分类，但是分类效果不好，使用三维的二分类网络进行分类效果较好。
参照图2，在一实施例中，上述通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像的步骤S1之前，包括：
步骤S101，对所述肺部CT图像进行预处理，以去除图像噪音。
由于医院使用医疗设备拍摄的病人肺部CT图像，具有较多的噪音，例如骨头的亮斑，CT床的金属线条等，因此，在对肺部CT图像进行分割之前，需要去除肺部CT图像中的噪音。该预处理的过程为传统方法，例如，在一具体实施例中，上述预处理过程可以是：使用-600HU作为阈值对肺部CT图像进行二值化处理，由于肺部CT图像的上下区域一般会有些切片与外界连接，需要去除；因此，将最终的图像像素值裁剪至[-1200，600]，再缩放至[0，255]；其中非肺部区域的像素点设置为170。
在其它实施例中，上述预处理过程还可以对肺部CT图像的像素间隔、图像对比度等进行调整。
在一实施例中，所述通过三维的卷积神经网络分割模型对肺部CT图像进行分割的步骤S1，包括：
使用多组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行卷积，以及对卷积结果进行上采样处理以获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
在本实施例中，上述三维的卷积神经网络的分割模型，其结构上具有类似于VGG网络（一种卷积神经网络）的多组卷积层，并加入了batch normalization（批规范化）层对每个卷积层进行卷积和上采样。其中，上采样是下采样的逆过程，上采样和下采样都是对数字信号进行重采，重采的采样率与原来获得该数字信号(比如从模拟信号采样而来)的采样率比较，大于原信号的称为上采样，小于的则称为下采样，上采样的实质也就是内插或插值。
在一具体实施例中，使用5组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行1X1X1的卷积，以及对卷积结果进行上采样处理之后，获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
在一实施例中，上述步骤S2中的三维的U-Net检测模型使用的损失函数为focal loss函数以及回归损失函数。使用focal loss函数的目的是解决检测模型训练过程中正负样本严重不均衡的问题。简单的负样本与难例的负样本对模型的贡献相同，focal loss可以使模型把注意力放在难例的负样本区分上面，使模型结果在FROC曲线上有更好的表现。
在一实施例中，上述通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节的步骤S2，包括：
a、对所述肺部区域图像依次经过四次convolution和max pooling（卷积和池化），以及两次deconvolution（逆卷积）计算得到第一概率图；并在所述肺部区域图像经过两次deconvolution计算之前分别加入一个分支，两个所述分支分别经过对应的deconvolution计算得到一个相应的第二概率图；
b、将两个所述第二概率图以及第一概率图同时输入至反向传播算法中进行迭代计算，得出最终概率图，所述最终概率图表示所述肺部区域图像中肺结节的概率；
c、根据所述最终概率图，从所述肺部区域图像中检测出可疑结节。
本实施例中使用的是三维的U-Net检测模型，限于GPU（图形处理器）存储大小，需要对上述分割出的肺部区域图像进行样本处理，具体地，输入图像时选择输入128*128*128的正方体，此时，70%的输入正方体中拥有一个结节，剩下的30%采取随机裁剪，使样本中包含背景样本。大型肺结节相较于小型肺结节而言数量较少，所以在去样本的时候，对于直径大于30mm，40mm的结节样本扩充2-6倍。同时，上述的所有样本都需要按概率翻转缩放以消除过拟合问题。
如上述步骤a所述，在本实施例中，将上述进行过样本处理的128*128*128正方体图像输入至三维的U-Net检测模型中，经过4次convolution和max pooling计算之后，再经过2次deconvolution计算得到32*32*32的第一概率图；同时在每个deconvolution计算之前加入新的分支，每个分支分别经过相应的deconvolution层计算输出32*32*32的第二概率图；如步骤b所述，将两个32*32*32的第二概率图以及32*32*32的第一概率图同时输入至反向传播算法中进行迭代计算，得出最终概率图，所述最终概率图表示所述肺部区域图像中肺结节的概率；根据所述最终概率图，从所述肺部区域图像中检测出可疑结节；概率超出预设值，则将其作为可疑结节。
具体地，检测可疑结节时，需要识别前景与背景，在本实施例中，使用IOU（Intersection-over-Union，交并比）指标决定前景与背景，IOU是对分割模型的一种评价指标，是检测区域和真实区域的交集与并集的比值，所得的值越大代表分割模型的分割结果与真实值越接近。这里前景是指目标，即肺结节，IOU大于0.5的为前景，小于0.02的是背景。训练分割模型时，我们使用IOU这个指标来确定当前检测窗口是否包含肺结节，即当前检测窗口与真实肺结节的IOU，大于0.5时认为当前检测窗口包含肺结节，小于0.02的被当做负样本，即背景，在这0.02-0.5之间的样本被舍去。由于肺结节的大小分布在3mm-30mm这个较大的区间，需要使用不同大小的检测窗口大小来检测肺结节。因此，本实施例中，检测模型中使用了5种anchor（锚点）：4、6、12、20、30。anchor的值用来定义检测窗口的大小，比如anchor=4则是以4x4x4的检测窗口大小来检测可疑结节。
在一实施例中，上述通过三维的二分类网络对所述可疑结节进行分类，去除假结节的步骤S3，具体包括：
S301，将所述可疑结节输入至三维的二分类网络的输入层，依次经过五个卷积层得到特征图；
S302，将所述特征图依次输入至三个全连接层，经所述全连接层输入至softmax层，经过所述softmax层输出得到所述可疑结节分类的概率，所述可疑结节分类的概率为可疑结节检测为肺结节的置信度；
S303，去除假结节，所述假结节为置信度低于设定值的可疑结节。
具体地，所述将所述特征图依次输入至三个全连接层的步骤，具体包括：
将所述特征图输入至batch normalization层，并经所述batch normalization层依次输入至三个全连接层。
在本实施例中，所述三维的二分类网络的网络结构中包括5个卷积层，三个全连接层，以及一个batch normalization层以及softmax层。本实施例中，如S301所述，将步骤S2中检测出的可疑结节以36X36X20大小的立方体输入至三维的二分类网络中，经过5个卷积层得到特征图，如S302所述，将特征图输入至batch normalization层之后，再依次输入至三个全连接层，最后经过softmax输出得到所述可疑结节分类的概率，该可疑结节分类的概率作为可疑结节检测为肺结节的置信度。预先设置有一个设定值，当置信度高于该设定值，则判断其为肺结节；若置信度低于上述设定值，则判断其为假结节，即假阳性，需要去除，以达到抑制假阳性的目的，从而保证本实施例中检测肺结节的检出率，降低医生工作量。
在另一实施例中，经过上述实施例的检测过程检测出肺部CT图像中的肺结节之后，则可以确定肺结节位置，并在三维图像中描绘出肺结节的三维形状，计算出肺结节的大小，将上述监测出的信息发送至医用电脑设备上进行显示，以便医生根据上述信息作出合理的治疗方案。由于肺结节的大小、形状、位置等信息的不同，医生对其的治疗方案将有所不同，因此，获取上述信息，有利于医生作出合理的治疗方案。最终，还可以将病人的上述病症以及对应的治疗方案记录在数据库中进行保存。
在又一实施例中，检测出肺结节之后，确定肺结节位置、肺结节形状及大小，根据上述信息去历史诊断的数据库中匹配出相似/相近的案例，以便参考便于医生进行准确的诊断，对相似案例进行分析，还可以方便对该肺结节疾病进行分析。
综上所述，为本申请实施例中提供的肺结节的检测方法，通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像，分割速度快，使得后续检测出肺结节的速度加快；同时，可适用于对所有肺部CT图像进行肺部区域的分割；通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节，以及通过三维的二分类网络对所述可疑结节进行分类，去除假结节，提高检测肺结节的效果，以及提升检测准确率。
参照图3，本申请实施例中还提供了一种肺结节的检测装置，包括：
分割单元10，用于通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；
检测单元20，用于通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；
分类单元30，用于通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
在本实施例中，医院使用医疗设备拍摄病人肺部的CT图像，该肺部CT图像为三维图像。肺部CT图像中不仅包括肺部区域图像，还包括肺部周围的一些其它组织或者医疗设备图像等，其都可能对后续肺结节的检测过程造成干扰，同时也容易增大检测肺结节时的搜索区域。因此，在从病人的肺部CT图像中检测肺结节时，为了减少其它组织的干扰和不必要的搜索区域，通常分割单元10需要先对肺部CT图像进行分割，从中分割出肺部区域图像。现有技术中通常使用图像分割算法（例如阈值分割、区域分割、边缘分割、直方图法等）从肺部CT图像中分割出肺部区域图像，但是，图像分割算法不仅分割速度慢，而且无法适用于对所有的肺部CT图像进行分割。
通过二维的分割模型也可以实现肺部区域图像的分割，但是分割效果不理想。因此，本实施例的分割单元10中使用了三维的全卷积神经网络的分割模型对肺部CT图像进行分割来实现三维分割，从肺部CT图像中分割出肺部区域图像，减少其它组织的干扰和不必要的搜索区域。三维的卷积神经网络分割模型相对于二维的分割模型，其区别在于，二维分割模型将三维的肺部CT图像分割我二维图像，必然造成数据表达不全面，影响分割效果，且二维分割时需要从多个维度进行分割，造成分割速度较慢；而使用三维的卷积神经网络分割模型直接实现三维分割，分割出来的肺部区域图像还是三维图像，数据表达全面，分割效果好，分割速度快。
因此，在分割单元10中使用三维的卷积神经网络分割模型进行肺部区域图像的分割，相比于使用现有图像分割算法来分割具有两大优势：
①、分割速度更快，传统图像算法对每一个肺部CT图像进行分割需要使用2-8分钟的时间，而使用三维的卷积神经网络分割模型只需要5-10秒。
②、传统图像分割算法不能对所有的肺部CT图像分割出肺部区域（例如一些较为特殊的肺部CT图像则无法进行分割），严重阻碍了后续的检测过程，而使用三维的卷积神经网络分割模型则能保证对所有肺部CT图像进行分割处理。
在经过上述分割单元10的分割之后，则从肺部CT图像中分割出肺部区域图像；检测单元20则通过三维的U-Net检测模型从上述肺部区域图像中检测出可疑结节，可疑结节指的是疑似为肺结节的特征。U-Net是一种现有的针对二维图像的分割模型，当前还有一些其它二维的检测模型，比如faster-rcnn，ssd等也可以用于检测出可疑结节；由于上述肺部区域图像是三维数据，显然，使用三维的检测模型检测可疑结节的效果最好；因此，本实施例中使用三维的U-Net检测模型从中检测可疑结节。经过试验对比，本申请中使用三维的U-Net检测模型进行可疑结节的检测想对于其它检测模型的效果更好。
上述检测单元20检测出的可疑结节中包括有许多假结节（假阳性），若不去除这些假结节，将对医生带来大量不必要的工作，因此，为了保证较高的肺结节检出率，则需要对可疑结节进行进一步地筛选。
本实施例的分类单元30使用三维的二分类网络对所述可疑结节进行进一步地分类，得到每个可疑结节的置信度，去掉置信度较低的作为假结节（即去除假结节），只保留置信度较高的作为真结节，以此达到抑制假阳性的目的，提高肺结节检出率。在步骤S3中，也可以使用其它二维的分类模型进行分类，但是分类效果不好，使用三维的二分类网络进行分类效果较好。
在一实施例中，上述肺结节的检测装置还包括：
预处理单元101，用于对所述肺部CT图像进行预处理，以去除图像噪音。
由于医院使用医疗设备拍摄的病人肺部CT图像，具有较多的噪音，例如骨头的亮斑，CT床的金属线条等，因此，在对肺部CT图像进行分割之前，需要去除肺部CT图像中的噪音。该预处理的过程为传统方法，例如，在一具体实施例中，上述预处理过程可以是：使用-600HU作为阈值对肺部CT图像进行二值化处理，由于肺部CT图像的上下区域一般会有些切片与外界连接，需要去除；因此，将最终的图像像素值裁剪至[-1200，600]，再缩放至[0，255]；其中非肺部区域的像素点设置为170。
在其它实施例中，上述预处理单元101的预处理过程中还可以对肺部CT图像的像素间隔、图像对比度等进行调整。
在一实施例中，上述分割单元10具体用于：
使用多组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行卷积，以及对卷积结果进行上采样处理以获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
在本实施例中，上述三维的卷积神经网络的分割模型，其结构上具有类似于VGG网络的多组卷积层，并加入了batch normalization（批规范化）层对每个卷积层进行卷积和上采样。其中，上采样是下采样的逆过程，上采样和下采样都是对数字信号进行重采，重采的采样率与原来获得该数字信号(比如从模拟信号采样而来)的采样率比较，大于原信号的称为上采样，小于的则称为下采样，上采样的实质也就是内插或插值。
在一具体实施例中，使用5组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行1X1X1的卷积，以及对卷积结果进行上采样处理之后，获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
在一实施例中，所述检测单元20中的三维的U-Net检测模型使用的损失函数为focal loss函数以及回归损失函数。使用focal loss函数的目的是解决检测模型训练过程中正负样本严重不均衡的问题。简单的负样本与难例的负样本对模型的贡献相同，focal loss可以使模型把注意力放在难例的负样本区分上面，使模型结果在FROC曲线上有更好的表现。
参照图4，在一实施例中，上述检测单元20包括：
第一计算模块201，用于对所述肺部区域图像依次经过四次convolution和max pooling，以及两次deconvolution计算得到第一概率图；并在所述肺部区域图像经过两次deconvolution计算之前分别加入一个分支，两个所述分支分别经过对应的deconvolution计算得到一个相应的第二概率图；
第二计算模块202，用于将两个所述第二概率图以及第一概率图同时输入至反向传播算法中进行迭代计算，得出最终概率图，所述最终概率图表示所述肺部区域图像中肺结节的概率；
检测模块203，用于根据所述最终概率图，从所述肺部区域图像中检测出可疑结节。
本实施例中使用的是三维的U-Net检测模型，限于GPU存储大小，需要对上述分割出的肺部区域图像进行样本处理，具体地，输入图像时选择输入128*128*128的正方体，此时，70%的输入正方体中拥有一个结节，剩下的30%采取随机裁剪，使样本中包含背景样本。大型肺结节相较于小型肺结节而言数量较少，所以在去样本的时候，对于直径大于30mm，40mm的结节样本扩充2-6倍。同时，上述的所有样本都需要按概率翻转缩放以消除过拟合问题。
在本实施例中，第一计算模块201将上述进行过样本处理的128*128*128正方体图像输入至三维的U-Net检测模型中，经过4次convolution和max pooling计算之后再经过2次deconvolution计算得到32*32*32的第一概率图；同时在每个deconvolution计算之前加入新的分支，每个分支分别经过相应的deconvolution层计算输出32*32*32的第二概率图；第二计算模块202将两个32*32*32的第二概率图以及32*32*32的第一概率图同时输入至反向传播算法中进行迭代计算，得出最终概率图，所述最终概率图表示所述肺部区域图像中肺结节的概率；根据所述最终概率图，从所述肺部区域图像中检测出可疑结节。
具体地，检测模块203检测可疑结节时，需要识别前景与背景，在本实施例中，使用IOU指标决定前景与背景，IOU是对分割模型的一种评价指标，是检测区域和真实区域的交集与并集的比值，所得的值越大代表分割模型的分割结果与真实值越接近。这里前景是指目标，即肺结节，IOU大于0.5的为前景，小于0.02的是背景。训练分割模型时，我们使用IOU这个指标来确定当前检测窗口是否包含肺结节，即当前检测窗口与真实肺结节的IOU，大于0.5时认为当前检测窗口包含肺结节，小于0.02的被当做负样本，即背景，在这0.02-0.5之间的样本被舍去。由于肺结节的大小分布在3mm-30mm这个较大的区间，需要使用不同大小的检测窗口大小来检测肺结节。因此，本实施例中，检测模型中使用了5种anchor（锚点）：4、6、12、20、30。anchor的值用来定义检测窗口的大小，比如anchor=4则是以4x4x4的检测窗口大小来检测可疑结节。
参照图5，在一实施例中，上述分类单元30包括：
卷积模块301，用于将所述可疑结节输入至三维的二分类网络的输入层，依次经过五个卷积层得到特征图；
输出模块302，用于将所述特征图依次输入至三个全连接层，再经所述全连接层输入至softmax层，最后经过所述softmax层输出得到所述可疑结节分类的概率，所述可疑结节分类的概率为可疑结节检测为肺结节的置信度；
去除模块303，用于去除假结节，所述假结节为置信度低于设定值的可疑结节。
具体中，所述输出模块302将所述特征图依次输入至三个全连接层具体包括：
将所述特征图输入至batch normalization层，并经所述batch normalization层依次输入至三个全连接层。
在本实施例中，所述三维的二分类网络的网络结构中包括5个卷积层，三个全连接层，以及一个batch normalization层以及softmax层。本实施例中，卷积模块301将检测出的可疑结节以36X36X20大小的立方体输入至三维的二分类网络中，经过5个卷积层得到特征图，输出模块302将特征图输入至batch normalization层之后，再依次输入至三个全连接层，最后经过softmax输出得到所述可疑结节分类的概率，该可疑结节分类的概率作为可疑结节检测为肺结节的置信度。预先设置有一个设定值，当置信度高于该设定值，则判断其为肺结节；若置信度低于上述设定值，则判断其为假结节，即假阳性，需要通过去除模块303将其去除，以达到抑制假阳性的目的，从而保证本实施例中检测肺结节的检出率，降低医生工作量。
在另一实施例中，经过上述实施例的检测过程检测出肺部CT图像中的肺结节之后，则可以确定肺结节位置，并在三维图像中描绘出肺结节的三维形状，计算出肺结节的大小，将上述监测出的信息发送至医用电脑设备上进行显示，以便医生根据上述信息作出合理的治疗方案。由于肺结节的大小、形状、位置等信息的不同，医生对其的治疗方案将有所不同，因此，获取上述信息，有利于医生作出合理的治疗方案。最终，还可以将病人的上述病症以及对应的治疗方案记录在数据库中进行保存。
在又一实施例中，检测出肺结节之后，确定肺结节位置、肺结节形状及大小，根据上述信息去历史诊断的数据库中匹配出相似/相近的案例，以便参考便于医生进行准确的诊断，对相似案例进行分析，还可以方便对该肺结节疾病进行分析。
综上所述，为本申请实施例中提供的肺结节的检测装置，分割单元10通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像，分割速度快，使得后续检测出肺结节的速度加快；同时，可适用于对所有肺部CT图像进行肺部区域的分割；检测单元20通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节，以及分类单元30通过三维的二分类网络对所述可疑结节进行分类，去除假结节，提高检测肺结节的效果，以及提升检测准确率。
参照图6，本申请实施例中还提供一种计算机设备，该计算机设备可以是服务器，其内部结构可以如图6所示。该计算机设备包括通过系统总线连接的处理器、存储器、网络接口和数据库。其中，该计算机设计的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统、计算机可读指令和数据库。该内存储器为非易失性存储介质中的操作系统和计算机可读指令的运行提供环境。该计算机设备的数据库用于存储三维卷积神经网络分割模型等数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机可读指令被处理器执行时以实现一种肺结节的检测方法。
上述处理器执行上述肺结节的检测方法的步骤：通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
在一实施例中，所述处理器通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像的步骤之前，包括：
对所述肺部CT图像进行预处理，以去除图像噪音。
在一实施例中，所述处理器通过三维的卷积神经网络分割模型对肺部CT图像进行分割的步骤，包括：
使用多组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行卷积，以及对卷积结果进行上采样处理以获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
在一实施例中，所述三维的U-Net检测模型使用的损失函数为focal loss函数以及回归损失函数。
在一实施例中，所述处理器通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节的步骤，包括：
对所述肺部区域图像依次经过四次convolution和max pooling，以及两次deconvolution计算得到第一概率图；并在所述肺部区域图像经过两次deconvolution计算之前分别加入一个分支，两个所述分支分别经过对应的deconvolution计算得到一个相应的第二概率图；
将两个所述第二概率图以及第一概率图同时输入至反向传播算法中进行迭代计算，得出最终概率图，所述最终概率图表示所述肺部区域图像中肺结节的概率；
根据所述最终概率图，从所述肺部区域图像中检测出可疑结节。
在一实施例中，所述处理器通过三维的二分类网络对所述可疑结节进行分类，去除假结节的步骤，包括：
将所述可疑结节输入至三维的二分类网络的输入层，依次经过五个卷积层得到特征图；
将所述特征图依次输入至三个全连接层，再经所述全连接层输入至softmax层，最后经过所述softmax层输出得到所述可疑结节分类的概率，所述可疑结节分类的概率为可疑结节检测为肺结节的置信度；
去除假结节，所述假结节为置信度低于设定值的可疑结节。
在一实施例中，所述处理器将所述特征图依次输入至三个全连接层的步骤，具体包括：
将所述特征图输入至batch normalization层，并经所述batch normalization层依次输入至三个全连接层。
本领域技术人员可以理解，图6中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定。
本申请一实施例还提供一种计算机非易失性可读存储介质，其上存储有计算机可读指令，计算机可读指令被处理器执行时实现一种肺结节的检测方法，具体为：通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
在一实施例中，所述处理器通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像的步骤之前，包括：
对所述肺部CT图像进行预处理，以去除图像噪音。
在一实施例中，所述处理器通过三维的卷积神经网络分割模型对肺部CT图像进行分割的步骤，包括：
使用多组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行卷积，以及对卷积结果进行上采样处理以获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
在一实施例中，所述三维的U-Net检测模型使用的损失函数为focal loss函数以及回归损失函数。
在一实施例中，所述处理器通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节的步骤，包括：
对所述肺部区域图像依次经过四次convolution和max pooling，以及两次deconvolution计算得到第一概率图；并在所述肺部区域图像经过两次deconvolution计算之前分别加入一个分支，两个所述分支分别经过对应的deconvolution计算得到一个相应的第二概率图；
将两个所述第二概率图以及第一概率图同时输入至反向传播算法中进行迭代计算，得出最终概率图，所述最终概率图表示所述肺部区域图像中肺结节的概率；
根据所述最终概率图，从所述肺部区域图像中检测出可疑结节。
在一实施例中，所述处理器通过三维的二分类网络对所述可疑结节进行分类，去除假结节的步骤，包括：
将所述可疑结节输入至三维的二分类网络的输入层，依次经过五个卷积层得到特征图；
将所述特征图依次输入至三个全连接层，再经所述全连接层输入至softmax层，最后经过所述softmax层输出得到所述可疑结节分类的概率，所述可疑结节分类的概率为可疑结节检测为肺结节的置信度；
去除假结节，所述假结节为置信度低于设定值的可疑结节。
在一实施例中，所述处理器将所述特征图依次输入至三个全连接层的步骤，具体包括：
将所述特征图输入至batch normalization层，并经所述batch normalization层依次输入至三个全连接层。
综上所述，为本申请实施例中提供的肺结节的检测方法、装置、计算机设备和存储介质，通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像，分割速度快，使得后续检测出肺结节的速度加快；同时，可适用于对所有肺部CT图像进行肺部区域的分割；通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节，以及通过三维的二分类网络对所述可疑结节进行分类，去除假结节，提高检测肺结节的效果，以及提升检测准确率。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储与一非易失性计算机可读取存储介质中，该计算机可读指令在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的和实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和/或易失性存储器。非易失性存储器可以包括只读存储器（ROM）、可编程ROM（PROM）、电可编程ROM（EPROM）、电可擦除可编程ROM（EEPROM）或闪存。易失性存储器可包括随机存取存储器（RAM）或者外部高速缓冲存储器。作为说明而非局限，RAM通过多种形式可得，诸如静态RAM（SRAM）、动态RAM（DRAM）、同步DRAM（SDRAM）、双速据率SDRAM（SSRSDRAM）、增强型SDRAM（ESDRAM）、同步链路（Synchlink）DRAM（SLDRAM）、存储器总线（Rambus）直接RAM（RDRAM）、直接存储器总线动态RAM（DRDRAM）、以及存储器总线动态RAM（RDRAM）等。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其它变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、装置、物品或者方法不仅包括那些要素，而且还包括没有明确列出的其它要素，或者是还包括为这种过程、装置、物品或者方法所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、装置、物品或者方法中还存在另外的相同要素。
以上所述仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其它相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种肺结节的检测方法，其特征在于，包括以下步骤：
    通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；
    通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；
    通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
  2. 根据权利要求1所述的肺结节的检测方法，其特征在于，所述通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像的步骤之前，包括：
    对所述肺部CT图像进行预处理，以去除图像噪音。
  3. 根据权利要求1所述的肺结节的检测方法，其特征在于，所述通过三维的卷积神经网络分割模型对肺部CT图像进行分割的步骤，包括：
    使用多组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行卷积，以及对卷积结果进行上采样处理以获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
  4. 根据权利要求1所述的肺结节的检测方法，其特征在于，所述三维的U-Net检测模型使用的损失函数为focal loss函数以及回归损失函数。
  5. 根据权利要求1所述的肺结节的检测方法，其特征在于，所述通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节的步骤，包括：
    对所述肺部区域图像依次经过四次convolution和max pooling，以及两次deconvolution计算得到第一概率图；并在所述肺部区域图像经过两次deconvolution计算之前分别加入一个分支，两个所述分支分别经过对应的deconvolution计算得到一个相应的第二概率图；
    将两个所述第二概率图以及第一概率图同时输入至反向传播算法中进行迭代计算，得出最终概率图，所述最终概率图表示所述肺部区域图像中肺结节的概率；
    根据所述最终概率图，从所述肺部区域图像中检测出可疑结节。
  6. 根据权利要求1所述的肺结节的检测方法，其特征在于，所述通过三维的二分类网络对所述可疑结节进行分类，去除假结节的步骤，包括：
    将所述可疑结节输入至三维的二分类网络的输入层，依次经过五个卷积层得到特征图；
    将所述特征图依次输入至三个全连接层，再经所述全连接层输入至softmax层，最后经过所述softmax层输出得到所述可疑结节分类的概率，所述可疑结节分类的概率为可疑结节检测为肺结节的置信度；
    去除假结节，所述假结节为置信度低于设定值的可疑结节。
  7. 根据权利要求6所述的肺结节的检测方法，其特征在于，所述将所述特征图依次输入至三个全连接层的步骤，具体包括：
    将所述特征图输入至batch normalization层，并经所述batch normalization层依次输入至三个全连接层。
  8. 一种肺结节的检测装置，其特征在于，包括：
    分割单元，用于通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；
    检测单元，用于通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；
    分类单元，用于通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
  9. 根据权利要求8所述的肺结节的检测装置，其特征在于，还包括：
    预处理单元，用于对所述肺部CT图像进行预处理，以去除图像噪音。
  10. 根据权利要求8所述的肺结节的检测装置，其特征在于，所述分割单元具体用于：
    使用多组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行卷积，以及对卷积结果进行上采样处理以获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
  11. 根据权利要求8所述的肺结节的检测装置，其特征在于，所述三维的U-Net检测模型使用的损失函数为focal loss函数以及回归损失函数。
  12. 根据权利要求8所述的肺结节的检测装置，其特征在于，所述检测单元包括：
    第一计算模块，用于对所述肺部区域图像依次经过四次convolution和max pooling，以及两次deconvolution计算得到第一概率图；并在所述肺部区域图像经过两次deconvolution计算之前分别加入一个分支，两个所述分支分别经过对应的deconvolution计算得到一个相应的第二概率图；
    第二计算模块，用于将两个所述第二概率图以及第一概率图同时输入至反向传播算法中进行迭代计算，得出最终概率图，所述最终概率图表示所述肺部区域图像中肺结节的概率；
    检测模块，用于根据所述最终概率图，从所述肺部区域图像中检测出可疑结节。
  13. 根据权利要求8所述的肺结节的检测装置，其特征在于，所述分类单元包括：
    卷积模块，用于将所述可疑结节输入至三维的二分类网络的输入层，依次经过五个卷积层得到特征图；
    输出模块，用于将所述特征图依次输入至三个全连接层，再经所述全连接层输入至softmax层，最后经过所述softmax层输出得到所述可疑结节分类的概率，所述可疑结节分类的概率为可疑结节检测为肺结节的置信度；
    去除模块，用于去除假结节，所述假结节为置信度低于设定值的可疑结节。
  14. 根据权利要求13所述的肺结节的检测装置，其特征在于，所述输出模块具体用于：
    将所述特征图输入至batch normalization层，并经所述batch normalization层依次输入至三个全连接层。
  15. 一种计算机设备，包括存储器和处理器，所述存储器存储有计算机可读指令，其特征在于，所述处理器执行所述计算机可读指令时实现肺结节的检测方法，所述方法包括：
    通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；
    通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；
    通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
  16. 根据权利要求15所述的计算机设备，其特征在于，所述处理器通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像的步骤之前，包括：
    对所述肺部CT图像进行预处理，以去除图像噪音。
  17. 根据权利要求15所述的计算机设备，其特征在于，所述处理器通过三维的卷积神经网络分割模型对肺部CT图像进行分割的步骤，包括：
    使用多组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行卷积，以及对卷积结果进行上采样处理以获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
  18. 一种计算机非易失性可读存储介质，其上存储有计算机可读指令，其特征在于，所述计算机可读指令被处理器执行时实现肺结节的检测方法，所述方法包括：
    通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像；
    通过三维的U-Net检测模型从所述肺部区域图像中检测出可疑结节；
    通过三维的二分类网络对所述可疑结节进行分类，去除假结节。
  19. 根据权利要求18所述的计算机非易失性可读存储介质，其特征在于，所述处理器通过三维的卷积神经网络分割模型对肺部CT图像进行分割，分割出肺部区域图像的步骤之前，包括：
    对所述肺部CT图像进行预处理，以去除图像噪音。
  20. 根据权利要求18所述的计算机非易失性可读存储介质，其特征在于，所述处理器通过三维的卷积神经网络分割模型对肺部CT图像进行分割的步骤，包括：
    使用多组卷积层提取所述肺部CT图像中的肺部区域特征，并在每组卷积层中加入批规范化方法对每组卷积层进行卷积，以及对卷积结果进行上采样处理以获得与所述肺部CT图像原始尺寸大小一致的肺部区域图像。
PCT/CN2018/095459 2018-04-20 2018-07-12 肺结节的检测方法、装置、计算机设备和存储介质 Ceased WO2019200740A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810362199.0 2018-04-20
CN201810362199.0A CN108765369B (zh) 2018-04-20 2018-04-20 肺结节的检测方法、装置、计算机设备和存储介质

Publications (1)

Publication Number Publication Date
WO2019200740A1 true WO2019200740A1 (zh) 2019-10-24

Family

ID=64011194

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/095459 Ceased WO2019200740A1 (zh) 2018-04-20 2018-07-12 肺结节的检测方法、装置、计算机设备和存储介质

Country Status (2)

Country Link
CN (1) CN108765369B (zh)
WO (1) WO2019200740A1 (zh)

Cited By (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110288610A (zh) * 2019-06-05 2019-09-27 苏州比格威医疗科技有限公司 一种视网膜oct硬性渗出分割方法
CN110992312A (zh) * 2019-11-15 2020-04-10 上海联影智能医疗科技有限公司 医学图像处理方法、装置、存储介质及计算机设备
CN111079821A (zh) * 2019-12-12 2020-04-28 哈尔滨市科佳通用机电股份有限公司 一种脱轨自动制动拉环脱落故障图像识别方法
CN111080605A (zh) * 2019-12-12 2020-04-28 哈尔滨市科佳通用机电股份有限公司 一种铁路货车人力制动机轴链脱落故障图像识别方法
CN111091564A (zh) * 2019-12-25 2020-05-01 金华市中心医院 一种基于3DUnet的肺结节大小检测系统
CN111160442A (zh) * 2019-12-24 2020-05-15 上海联影智能医疗科技有限公司 图像分类方法、计算机设备和存储介质
CN111166070A (zh) * 2019-12-17 2020-05-19 五邑大学 一种基于指静脉认证的医疗储物柜及其管理方法
CN111340756A (zh) * 2020-02-13 2020-06-26 北京深睿博联科技有限责任公司 一种医学图像病变检出合并方法、系统、终端及存储介质
CN111402231A (zh) * 2020-03-16 2020-07-10 杭州健培科技有限公司 一种用于肺部ct影像质量的自动评估系统及方法
CN111476766A (zh) * 2020-03-31 2020-07-31 哈尔滨商业大学 基于深度学习的肺结节ct图像检测系统
CN111696084A (zh) * 2020-05-20 2020-09-22 平安科技（深圳）有限公司 细胞图像分割方法、装置、电子设备及可读存储介质
CN111754453A (zh) * 2020-05-11 2020-10-09 佛山市第四人民医院(佛山市结核病防治所) 基于胸透图像的肺结核检测方法、系统和存储介质
CN111754532A (zh) * 2020-08-12 2020-10-09 腾讯科技（深圳）有限公司 图像分割模型搜索方法、装置、计算机设备及存储介质
CN111784638A (zh) * 2020-06-04 2020-10-16 广东省智能制造研究所 一种基于卷积神经网络的肺结节假阳性筛除方法及系统
CN112184657A (zh) * 2020-09-24 2021-01-05 上海健康医学院 一种肺结节自动检测方法、装置及计算机系统
CN112258461A (zh) * 2020-10-13 2021-01-22 江南大学 一种基于卷积神经网络的肺结节检测方法
CN112365504A (zh) * 2019-10-29 2021-02-12 杭州脉流科技有限公司 Ct左心室分割方法、装置、设备和存储介质
CN112508057A (zh) * 2020-11-13 2021-03-16 上海健康医学院 一种肺结节分类方法、介质及电子设备
CN113222024A (zh) * 2021-05-17 2021-08-06 点内（上海）生物科技有限公司 一种基于深度学习的肺部疾病多层次分类方法、系统及存储介质
CN113554581A (zh) * 2020-04-21 2021-10-26 北京灵汐科技有限公司 一种三维医学影像识别方法和系统
CN113902676A (zh) * 2021-09-08 2022-01-07 山东师范大学 基于注意力机制的神经网络的肺结节图像检测方法及系统
CN114119571A (zh) * 2021-11-30 2022-03-01 湖南大学 一种基于双路三维卷积神经网络的肺结节检测方法
CN114283165A (zh) * 2021-12-17 2022-04-05 上海交通大学 肺结节的智能影像处理系统
CN114529516A (zh) * 2022-01-17 2022-05-24 重庆邮电大学 基于多注意力与多任务特征融合的肺结节检测与分类方法
CN115511818A (zh) * 2022-09-21 2022-12-23 北京医准智能科技有限公司 一种肺结节检出模型的优化方法、装置、设备及存储介质
CN115908286A (zh) * 2022-11-04 2023-04-04 武汉大学 一种多域肺结节检测方法、装置、设备及可读存储介质
CN116152168A (zh) * 2022-12-13 2023-05-23 深圳技术大学 一种医肺影像病变分类方法和分类装置
CN116681662A (zh) * 2023-05-26 2023-09-01 杭州华匠医学机器人有限公司 基于图像的肺结节术中定位方法、装置、电子设备及介质
CN116934888A (zh) * 2022-03-31 2023-10-24 富士胶片医疗健康株式会社 磁共振成像装置、图像处理装置以及图像处理方法
CN117274198A (zh) * 2023-09-22 2023-12-22 华中科技大学 一种基于多任务学习的肺结节检测和语义属性评级方法
CN117853806A (zh) * 2024-01-10 2024-04-09 天津市中心妇产科医院 妇科肿瘤图像处理系统及其方法
CN118968227A (zh) * 2024-10-12 2024-11-15 真健康(广东横琴)医疗科技有限公司 肺结节识别模型训练方法及设备、肺结节识别设备
US12171610B2 (en) 2019-10-22 2024-12-24 Shanghai United Imaging Intelligence Co., Ltd. Systems and methods for lung nodule evaluation

Families Citing this family (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109727253A (zh) * 2018-11-14 2019-05-07 西安大数据与人工智能研究院 基于深度卷积神经网络自动分割肺结节的辅助检测方法
CN109741823A (zh) * 2018-11-23 2019-05-10 杭州电子科技大学 一种基于深度学习的气胸辅助诊断方法
CN109685768B (zh) * 2018-11-28 2020-11-20 心医国际数字医疗系统(大连)有限公司 基于肺部ct序列的肺结节自动检测方法及系统
CN109697459A (zh) * 2018-12-04 2019-04-30 云南大学 一种面向光学相干断层图像斑块形态检测方法
CN109685776B (zh) * 2018-12-12 2021-01-19 华中科技大学 一种基于ct图像的肺结节检测方法及系统
CN109801294A (zh) * 2018-12-14 2019-05-24 深圳先进技术研究院 三维左心房分割方法、装置、终端设备及存储介质
CN109886967A (zh) * 2019-01-16 2019-06-14 成都蓝景信息技术有限公司 基于深度学习技术的肺部解剖学位置定位算法
CN110009599A (zh) * 2019-02-01 2019-07-12 腾讯科技（深圳）有限公司 肝占位检测方法、装置、设备及存储介质
CN109978004B (zh) * 2019-02-21 2024-03-29 平安科技（深圳）有限公司 图像识别方法及相关设备
CN109993726B (zh) * 2019-02-21 2021-02-19 上海联影智能医疗科技有限公司 医学图像的检测方法、装置、设备和存储介质
CN109919961A (zh) * 2019-02-22 2019-06-21 北京深睿博联科技有限责任公司 一种用于颅内cta图像中动脉瘤区域的处理方法及装置
CN109961443A (zh) * 2019-03-25 2019-07-02 北京理工大学 基于多期ct影像引导的肝脏肿瘤分割方法及装置
CN110059697B (zh) * 2019-04-29 2023-04-28 上海理工大学 一种基于深度学习的肺结节自动分割方法
CN110544256B (zh) * 2019-08-08 2022-03-22 北京百度网讯科技有限公司 基于稀疏特征的深度学习图像分割方法及装置
CN110599502B (zh) * 2019-09-06 2023-07-11 江南大学 一种基于深度学习的皮肤病变分割方法
CN110930414B (zh) * 2019-10-17 2024-10-22 平安科技（深圳）有限公司 医学影像的肺部区域阴影标记方法、装置、服务器及存储介质
CN110728675A (zh) * 2019-10-22 2020-01-24 慧影医疗科技（北京）有限公司 肺结节分析装置、模型训练方法、装置及分析设备
CN110766682A (zh) * 2019-10-29 2020-02-07 慧影医疗科技（北京）有限公司 肺结核定位筛查装置及计算机设备
CN111667458B (zh) * 2020-04-30 2023-09-01 杭州深睿博联科技有限公司 一种对平扫ct中的早急性脑梗死检测方法及装置
CN111738992B (zh) * 2020-06-04 2023-12-22 讯飞医疗科技股份有限公司 肺部病灶区域提取方法、装置、电子设备和存储介质
CN111932559B (zh) * 2020-08-26 2022-11-29 上海市公共卫生临床中心 基于深度学习的新冠肺炎肺部病灶区域分割系统
CN112184659B (zh) * 2020-09-24 2023-08-25 上海健康医学院 一种肺部图像处理方法、装置及设备
CN112801964B (zh) * 2021-01-20 2022-02-22 中国人民解放军总医院 肺部ct图像的多标签智能检测方法、装置、设备和介质
CN112967279A (zh) * 2021-04-02 2021-06-15 慧影医疗科技（北京）有限公司 一种检测肺结节的方法、装置、存储介质和电子设备
CN113643308B (zh) * 2021-08-31 2025-08-22 深圳平安医疗健康科技服务有限公司 肺部图像分割方法、装置、存储介质及计算机设备
CN113838026B (zh) * 2021-09-22 2024-02-02 中南大学 非小细胞肺癌检测方法、装置、计算机设备和存储介质
CN113920137B (zh) * 2021-10-14 2024-06-18 平安科技（深圳）有限公司 淋巴结转移预测方法、装置、设备及存储介质
CN113990432B (zh) * 2021-10-28 2025-09-19 北京来也网络科技有限公司 基于rpa和ai的影像报告推送方法、装置及计算设备
CN114187252B (zh) * 2021-12-03 2022-09-20 推想医疗科技股份有限公司 图像处理方法及装置、调整检测框的方法及装置
CN114494236A (zh) * 2022-02-16 2022-05-13 东华大学 基于过完备卷积神经网络的织物缺陷检测方法及系统
CN115423819B (zh) * 2022-07-20 2025-07-18 东北大学 肺血管的分割方法及装置、电子设备和存储介质
CN116721068A (zh) * 2023-05-30 2023-09-08 杭州华匠医学机器人有限公司 基于ct图像的肺结节分割方法，图像处理设备及可读介质
CN119228765B (zh) * 2024-09-26 2025-10-21 北京大学第一医院(北京大学第一临床医学院) 一种基于胸部ct预测肾上腺病变的方法、设备及程序产品

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106780460A (zh) * 2016-12-13 2017-05-31 杭州健培科技有限公司 一种用于胸部ct影像的肺结节自动检测系统
CN106940816A (zh) * 2017-03-22 2017-07-11 杭州健培科技有限公司 基于3d全连接卷积神经网络的ct图像肺结节检测系统
CN107016665A (zh) * 2017-02-16 2017-08-04 浙江大学 一种基于深度卷积神经网络的ct肺结节检测方法
CN107274402A (zh) * 2017-06-27 2017-10-20 北京深睿博联科技有限责任公司 一种基于胸部ct影像的肺结节自动检测方法及系统
CN107909581A (zh) * 2017-11-03 2018-04-13 杭州依图医疗技术有限公司 Ct影像的肺叶段分割方法、装置、系统、存储介质及设备

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106204587B (zh) * 2016-05-27 2019-01-08 浙江德尚韵兴图像科技有限公司 基于深度卷积神经网络和区域竞争模型的多器官分割方法
CN106504232B (zh) * 2016-10-14 2019-06-14 北京网医智捷科技有限公司 一种基于3d卷积神经网络的肺部结节自动检测系统
CN107590797B (zh) * 2017-07-26 2020-10-30 浙江工业大学 一种基于三维残差神经网络的ct影像肺结节检测方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106780460A (zh) * 2016-12-13 2017-05-31 杭州健培科技有限公司 一种用于胸部ct影像的肺结节自动检测系统
CN107016665A (zh) * 2017-02-16 2017-08-04 浙江大学 一种基于深度卷积神经网络的ct肺结节检测方法
CN106940816A (zh) * 2017-03-22 2017-07-11 杭州健培科技有限公司 基于3d全连接卷积神经网络的ct图像肺结节检测系统
CN107274402A (zh) * 2017-06-27 2017-10-20 北京深睿博联科技有限责任公司 一种基于胸部ct影像的肺结节自动检测方法及系统
CN107909581A (zh) * 2017-11-03 2018-04-13 杭州依图医疗技术有限公司 Ct影像的肺叶段分割方法、装置、系统、存储介质及设备

Cited By (43)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110288610A (zh) * 2019-06-05 2019-09-27 苏州比格威医疗科技有限公司 一种视网膜oct硬性渗出分割方法
US12171610B2 (en) 2019-10-22 2024-12-24 Shanghai United Imaging Intelligence Co., Ltd. Systems and methods for lung nodule evaluation
CN112365504A (zh) * 2019-10-29 2021-02-12 杭州脉流科技有限公司 Ct左心室分割方法、装置、设备和存储介质
CN110992312B (zh) * 2019-11-15 2024-02-27 上海联影智能医疗科技有限公司 医学图像处理方法、装置、存储介质及计算机设备
CN110992312A (zh) * 2019-11-15 2020-04-10 上海联影智能医疗科技有限公司 医学图像处理方法、装置、存储介质及计算机设备
CN111079821A (zh) * 2019-12-12 2020-04-28 哈尔滨市科佳通用机电股份有限公司 一种脱轨自动制动拉环脱落故障图像识别方法
CN111080605A (zh) * 2019-12-12 2020-04-28 哈尔滨市科佳通用机电股份有限公司 一种铁路货车人力制动机轴链脱落故障图像识别方法
CN111166070A (zh) * 2019-12-17 2020-05-19 五邑大学 一种基于指静脉认证的医疗储物柜及其管理方法
CN111160442B (zh) * 2019-12-24 2024-02-27 上海联影智能医疗科技有限公司 图像分类方法、计算机设备和存储介质
CN111160442A (zh) * 2019-12-24 2020-05-15 上海联影智能医疗科技有限公司 图像分类方法、计算机设备和存储介质
CN111091564B (zh) * 2019-12-25 2024-04-26 金华市中心医院 一种基于3DUnet的肺结节大小检测系统
CN111091564A (zh) * 2019-12-25 2020-05-01 金华市中心医院 一种基于3DUnet的肺结节大小检测系统
CN111340756A (zh) * 2020-02-13 2020-06-26 北京深睿博联科技有限责任公司 一种医学图像病变检出合并方法、系统、终端及存储介质
CN111340756B (zh) * 2020-02-13 2023-11-28 北京深睿博联科技有限责任公司 一种医学图像病变检出合并方法、系统、终端及存储介质
CN111402231B (zh) * 2020-03-16 2023-05-23 杭州健培科技有限公司 一种用于肺部ct影像质量的自动评估系统及方法
CN111402231A (zh) * 2020-03-16 2020-07-10 杭州健培科技有限公司 一种用于肺部ct影像质量的自动评估系统及方法
CN111476766A (zh) * 2020-03-31 2020-07-31 哈尔滨商业大学 基于深度学习的肺结节ct图像检测系统
CN111476766B (zh) * 2020-03-31 2023-08-22 哈尔滨商业大学 基于深度学习的肺结节ct图像检测系统
CN113554581A (zh) * 2020-04-21 2021-10-26 北京灵汐科技有限公司 一种三维医学影像识别方法和系统
CN111754453A (zh) * 2020-05-11 2020-10-09 佛山市第四人民医院(佛山市结核病防治所) 基于胸透图像的肺结核检测方法、系统和存储介质
CN111696084A (zh) * 2020-05-20 2020-09-22 平安科技（深圳）有限公司 细胞图像分割方法、装置、电子设备及可读存储介质
CN111696084B (zh) * 2020-05-20 2024-05-31 平安科技（深圳）有限公司 细胞图像分割方法、装置、电子设备及可读存储介质
CN111784638A (zh) * 2020-06-04 2020-10-16 广东省智能制造研究所 一种基于卷积神经网络的肺结节假阳性筛除方法及系统
CN111754532A (zh) * 2020-08-12 2020-10-09 腾讯科技（深圳）有限公司 图像分割模型搜索方法、装置、计算机设备及存储介质
CN111754532B (zh) * 2020-08-12 2023-07-11 腾讯科技（深圳）有限公司 图像分割模型搜索方法、装置、计算机设备及存储介质
CN112184657A (zh) * 2020-09-24 2021-01-05 上海健康医学院 一种肺结节自动检测方法、装置及计算机系统
CN112258461B (zh) * 2020-10-13 2024-04-09 江南大学 一种基于卷积神经网络的肺结节检测方法
CN112258461A (zh) * 2020-10-13 2021-01-22 江南大学 一种基于卷积神经网络的肺结节检测方法
CN112508057A (zh) * 2020-11-13 2021-03-16 上海健康医学院 一种肺结节分类方法、介质及电子设备
CN113222024A (zh) * 2021-05-17 2021-08-06 点内（上海）生物科技有限公司 一种基于深度学习的肺部疾病多层次分类方法、系统及存储介质
CN113902676A (zh) * 2021-09-08 2022-01-07 山东师范大学 基于注意力机制的神经网络的肺结节图像检测方法及系统
CN114119571A (zh) * 2021-11-30 2022-03-01 湖南大学 一种基于双路三维卷积神经网络的肺结节检测方法
CN114283165A (zh) * 2021-12-17 2022-04-05 上海交通大学 肺结节的智能影像处理系统
CN114529516A (zh) * 2022-01-17 2022-05-24 重庆邮电大学 基于多注意力与多任务特征融合的肺结节检测与分类方法
CN116934888A (zh) * 2022-03-31 2023-10-24 富士胶片医疗健康株式会社 磁共振成像装置、图像处理装置以及图像处理方法
CN115511818B (zh) * 2022-09-21 2023-06-13 北京医准智能科技有限公司 一种肺结节检出模型的优化方法、装置、设备及存储介质
CN115511818A (zh) * 2022-09-21 2022-12-23 北京医准智能科技有限公司 一种肺结节检出模型的优化方法、装置、设备及存储介质
CN115908286A (zh) * 2022-11-04 2023-04-04 武汉大学 一种多域肺结节检测方法、装置、设备及可读存储介质
CN116152168A (zh) * 2022-12-13 2023-05-23 深圳技术大学 一种医肺影像病变分类方法和分类装置
CN116681662A (zh) * 2023-05-26 2023-09-01 杭州华匠医学机器人有限公司 基于图像的肺结节术中定位方法、装置、电子设备及介质
CN117274198A (zh) * 2023-09-22 2023-12-22 华中科技大学 一种基于多任务学习的肺结节检测和语义属性评级方法
CN117853806A (zh) * 2024-01-10 2024-04-09 天津市中心妇产科医院 妇科肿瘤图像处理系统及其方法
CN118968227A (zh) * 2024-10-12 2024-11-15 真健康(广东横琴)医疗科技有限公司 肺结节识别模型训练方法及设备、肺结节识别设备

Also Published As

Publication number Publication date
CN108765369B (zh) 2023-05-02
CN108765369A (zh) 2018-11-06

Similar Documents

Publication Publication Date Title
WO2019200740A1 (zh) 肺结节的检测方法、装置、计算机设备和存储介质
CN114066884B (zh) 视网膜血管分割方法及装置、电子设备和存储介质
CN106940816B (zh) 基于3d全卷积神经网络的ct图像肺结节检测系统
WO2023193401A1 (zh) 点云检测模型训练方法、装置、电子设备及存储介质
WO2022063199A1 (zh) 一种肺结节自动检测方法、装置及计算机系统
CN110599500B (zh) 一种基于级联全卷积网络的肝脏ct图像的肿瘤区域分割方法及系统
WO2022037642A1 (zh) 一种临床图像中病变区域的检测和分类方法
CN109685768B (zh) 基于肺部ct序列的肺结节自动检测方法及系统
WO2021203795A1 (zh) 一种基于显著性密集连接扩张卷积网络的胰腺ct自动分割方法
CN110378381A (zh) 物体检测方法、装置和计算机存储介质
CN108288271A (zh) 基于三维残差网络的图像检测系统及方法
CN114663380A (zh) 一种铝材表面缺陷检测方法、存储介质及计算机系统
CN116152492B (zh) 基于多重注意力融合的医学图像分割方法
CN114494160B (zh) 一种基于完全融合集成网络候选框的骨折检测方法
Jo et al. Segmentation of the main vessel of the left anterior descending artery using selective feature mapping in coronary angiography
CN113724267B (zh) 一种乳腺超声图像肿瘤分割方法及装置
CN118279667A (zh) 一种面向皮肤镜图像的深度学习白癜风识别方法
CN116309295A (zh) 基于dwi影像的急性缺血性脑卒中aspects自动评分装置
CN113269734B (zh) 一种基于元学习特征融合策略的肿瘤图像检测方法及装置
Çınar et al. Brain stroke detection from CT images using transfer learning method
CN117876493A (zh) 用于对眼前节结构进行定位的方法、设备及存储介质
CN117522890A (zh) 图像分割方法、装置、计算机设备和存储介质
Chandranegara et al. Analysis of Pneumonia on Chest X-Ray Images Using Convolutional Neural Network Model iResNet-RS
CN115631194B (zh) 颅内动脉瘤识别检测的方法、装置、设备及介质
CN115661174A (zh) 基于流扭曲的表面缺陷区域分割方法、装置及电子设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18915248

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18915248

Country of ref document: EP

Kind code of ref document: A1