Medical image display device, medical image display method, and program

Abstract

To improve the visibility of a visual line direction in a medical image.SOLUTION: A medical image display device according to an embodiment includes an acquisition unit, a visual line detection unit, and a display control unit. The acquisition unit acquires a medical image including at least one eyeball of a subject. The visual line detection unit detects a visual line direction of the eyeball included in the medical image. The display control unit determines a display mode of the medical image on the basis of the visual line direction of the eyeball, and displays the medical image on the display unit in the determined display mode.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] FIELD Embodiments described herein generally relate to a medical image display device, a medical image display method, and a program for displaying medical images, such as images of a patient suspected of having a stroke. [Background technology]

[0002] Stroke is an example of a serious, life-threatening medical condition that may require emergency medical treatment. Typically, a non-contrast CT scan (NCCT) may be performed first in diagnosing stroke. NCCT scan results may be used to rule out hemorrhagic stroke as a cause. NCCT scan results may be used to rule out conditions that mimic stroke symptoms, such as epileptic seizures and brain tumors. NCCT scan results may be used to identify dense vessels, which may indicate a blood clot, and / or to identify ischemia.

[0003] A subsequent CT angiography (CTA), which combines a CT scan with the injection of a contrast agent, may be performed to confirm the initial diagnosis and / or to obtain further information to aid in treatment decisions.

[0004] One of the causes of ischemic stroke is, for example, the presence of large vessel occlusion, which is an acute blockage of the anterior and posterior circulation.

[0005] If a patient has a large vessel occlusion, thrombectomy may be an appropriate treatment. Mechanical thrombectomy aims to restore blood flow by removing the obstructing clot using a thrombectomy device delivered through an intravascular catheter. The patient may need to be transported to a hospital capable of performing thrombectomy to receive thrombectomy treatment. The process of determining whether a large vessel occlusion exists and whether thrombectomy is indicated may be urgent.

[0006] Determining which patients have large vessel occlusions and may be eligible for potentially life-saving thrombectomy therapy can form a critical part of the stroke clinical workflow. Evidence of the benefits of endovascular thrombectomy continues to spur the need to rapidly identify potentially eligible patients.

[0007] Determining whether a patient has a large vessel occlusion typically requires an intracranial vascular imaging study, according to American Heart Association (AHA) guidelines. There are several challenges associated with this: There may be a lack of expertise available to interpret the vascular study; The timing of contrast administration may be problematic or the study may be nondiagnostic; Some patients have contraindications to the iodinated contrast used in vascular studies; Not all hospitals routinely perform intracranial vascular studies.

[0008] The presence of a given imaging feature on a non-contrast study may indicate the presence of obstruction. Signs of the presence of obstruction may have high specificity but low sensitivity. One such imaging feature is the hyper-dense artery sign (HAS) on non-contrast computed tomography (NCCT). Another such imaging feature is the susceptibility vessel sign (SVS) on a T2* gradient recalled echo (GRE) magnetic resonance study (MRI).

[0009] Clinical triage scales may be used to indicate whether a large vessel occlusion is present. Examples of clinical scales that may be used for LVO triage include the Rapid Arterial Occlusion Evaluation (RACE) scale and the Cincinnati Prehospital Stroke Scale.

[0010] Studies have shown that ocular deviation can identify patients who are more likely to have large-vessel occlusion. Ocular deviation is included as a component in most clinical scales used in LVO triage (e.g., RACE and Cincinnati).

[0011] In some situations, non-contrast imaging features or clinical triage scales may not be sensitive and specific enough to identify all potentially eligible thrombectomy candidates.

[0012] Several currently available LVO triage methods identify CTA studies in which potential LVO is present. In such currently available LVO triage methods, LVO is directly indicated. [Prior art documents] [Patent documents]

[0013] [Patent Document 1] Patent Publication No. 2021-20054 Summary of the Invention [Problem to be solved by the invention]

[0014] One of the problems that the embodiments disclosed in this specification and the drawings aim to solve is to improve the visibility of the line of sight in medical images. However, the problems that the embodiments disclosed in this specification and the drawings aim to solve are not limited to the above problem. Problems corresponding to the configurations shown in the embodiments described below can also be considered as other problems. [Means for solving the problem]

[0015] A medical image display device according to an embodiment includes an acquisition unit, a gaze detection unit, and a display control unit. The acquisition unit acquires a medical image including at least one eyeball of a subject. The gaze detection unit detects the gaze direction of the eyeball included in the medical image. The display control unit determines a display mode of the medical image based on the gaze direction of the eyeball, and displays the medical image on a display unit in the determined display mode. [Brief explanation of the drawings]

[0016] Embodiments will now be described by way of example and not limitation and are illustrated in the following figures: [Figure 1] FIG. 1 is a schematic diagram of an apparatus according to an embodiment. [Figure 2] FIG. 2 is a flow chart outlining a method according to an embodiment. [Figure 3] FIG. 3 is a schematic diagram of a notification according to an embodiment. [Figure 4] FIG. 4 is a schematic diagram of a user interface according to an embodiment. [Figure 5] FIG. 5 is a schematic diagram of a user interface according to an embodiment. [Figure 6] FIG. 6 is a schematic diagram of a user interface according to an embodiment with persistent ocular deviation. [Figure 7] FIG. 7 is a schematic diagram of a user interface according to an embodiment with high absorption vessels. [Figure 8] FIG. 8 is a diagram of a patient's eye with gaze direction highlighted according to an embodiment. [Figure 9] FIG. 9 is a diagram of a patient's eye with gaze direction highlighted according to an embodiment. [Figure 10] FIG. 10 is a diagram of a patient's eye with gaze direction highlighted according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0017] One embodiment provides an image display device (medical image display device) comprising a processing circuit configured to receive medical image data including a representation of at least one eye of a subject, process the medical image data to determine a gaze direction of the at least one eye of the subject, and select a display manner for displaying the medical image data depending on the determined gaze direction.

[0018] One embodiment provides a method comprising receiving medical image data including a representation of at least one eye of a subject, processing the medical image data to determine a gaze direction of the at least one eye of the subject, and selecting a display manner for displaying the medical image data depending on the determined gaze direction.

[0019] A medical imaging device 10 according to an embodiment is shown schematically in Figure 1. The medical imaging device 10 is configured to process and display medical images of a patient or other subject. The medical imaging device 10 may also be referred to as an image display device or image display device (medical image display device).

[0020] The medical imaging device 10 comprises a computing device 12, which in this example is a personal computer (PC) or workstation. The computing device 12 is connected to a display device 16, e.g., a screen, and one or more input devices 18, e.g., a computer keyboard and a mouse. In some embodiments, the display device 16 is a touchscreen that also functions as the input device 18. The computing device 12 is connected to a data store 20.

[0021] The medical imaging device 10 is connected to a CT scanner 14 configured to perform non-contrast CT scans (NCCT) and CT angiography (CTA) scans of a patient or other subject to obtain volumetric medical imaging data. In this embodiment, each scan includes a scan of the brain. In other embodiments, any suitable body part may be scanned.

[0022] In alternative embodiments, data may be obtained using any suitable modality and / or acquisition method. CT scanner 14 may be substituted or supplemented by one or more scanners configured to obtain two-dimensional or three-dimensional imaging data using any suitable imaging modality, such as a CT scanner, a cone-beam CT scanner, a Magnetic Resonance Imaging (MRI) scanner, an X-ray scanner, an ultrasound scanner, a Positron Emission Tomography (PET) scanner, or a Single Photon Emission Computed Tomography (SPECT) scanner.

[0023] Data acquired using the CT scanner 14 is stored in a data store 20 and provided to the computing device 12. In other embodiments, the computing device 12 may acquire the data directly from the CT scanner 14. In alternative embodiments, the medical imaging device 10 receives medical imaging data and / or medical images from one or more additional data stores (not shown) instead of or in addition to the data store 20. For example, the medical imaging device 10 may receive medical imaging data from one or more remote data stores that may form part of a Picture Archiving and Communication System (PACS) or other information systems, such as, for example, a laboratory archive, an Electronic Medical Record (EMR) system, or an Admission Discharge and Transfer (ADT) system. Here, the medical imaging device 10, which implements the functionality of receiving medical imaging data and / or medical images, is an example of an acquisition unit.

[0024] Computing device 12 includes a Central Processing Unit (CPU) 22. Computing device 12 provides processing resources for automatically or semi-automatically processing a dataset, which in this embodiment comprises medical imaging data.

[0025] The computing device 12 includes a feature detection circuit 23 configured to process the imaging data to identify one or more imaging features, such as detecting regions of high-absorption vessels (thrombi), a gaze detection circuit 24 configured to determine a direction of gaze, a notification circuit 25 configured to flag potential thrombus removal candidates, a rendering circuit 26 configured to render an image from the imaging data, and a display circuit 28 configured to select and position the rendered view on the display screen 16 or any suitable display. Here, the feature detection circuit 23 is an example of a feature detection unit. Also, the gaze detection circuit 24 is an example of a gaze detection unit. Also, the notification circuit 25 is an example of a notification unit. Also, the rendering circuit 26 is an example of an image processing unit. Also, the display circuit 28 is an example of a display control unit. Also, the display device 16 (display screen 16) is an example of a display unit.

[0026] In this embodiment, circuits 23, 24, 25, 26, and 28 are each implemented on computing device 12 by a computer program having computer-readable instructions executable to perform the method of the embodiment. However, in other embodiments, various circuits may be implemented as one or more Application Specific Integrated Circuits (ASICs) or Field Programmable Gate Arrays (FPGAs). In this embodiment, circuits 23, 24, 25, 26, and 28 are each implemented as part of CPU 22. In alternative embodiments, circuits 23, 24, 25, 26, and 28 may be implemented separately or may form part of two or more CPUs. In further embodiments, at least a portion of the method may be performed on one or more Graphical Processing Units (GPUs).

[0027] Computing device 12 also has a hard drive and other components of a PC, including RAM, ROM, a data bus, an operating system including various device drivers, and hardware devices including a graphics card, although such components are not shown in FIG.

[0028] The system of FIG. 1 is configured to carry out a series of stages outlined in the flow chart of FIG.

[0029] At stage 30, the CT scanner acquires a non-contrast CT (NCCT) scan of the brain of a patient suspected of having a stroke. It is common clinical practice to perform a non-contrast CT scan of the brain when a stroke is suspected.

[0030] The volumetric data obtained by the NCCT scan is provided to a data store 20, and from there to a feature detection circuit 23 and a gaze detection circuit 24. In other embodiments, the feature detection circuit 23 and the gaze detection circuit 24 may obtain a set of volumetric NCCT data from any suitable data store. The NCCT data may be obtained by the feature detection circuit 23 and the gaze detection circuit 24 at any suitable time after the NCCT scan is performed. In these cases, the feature detection circuit 23 and the gaze detection circuit 24 may be considered examples of an acquisition unit.

[0031] At stage 32, the feature detection circuitry 23 performs a hyperdense vessel detection process. The hyperdense vessel detection process involves processing the volumetric NCCT data received at stage 30 to identify whether a hyperdense artery sign (HAS) is present in the NCCT scan. The HAS is sometimes also referred to as a hyperdense vessel sign.

[0032] In the embodiment of FIG. 2 , high-absorption blood vessel detection processing is performed using a convolutional neural network, for example, as described in the following method: Lisowska A., Beveridge E., Muir K. and Poole I. Thrombus Detection in CT Brain Scans using a Convolutional Neural Network. DOI: 10.5220 / 0006114600240033 In Proceedings of the 10th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2017), pages 24-33.

[0033] In other embodiments, any suitable high-absorption vessel detection method may be used, for example, any suitable image analysis method may be used to automatically detect high-absorption vessels, and relevant portions of the patient's anatomy may be displayed to the user.

[0034] In other embodiments, feature detection circuitry 23 may be configured to detect one or more non-contrast imaging features associated with an occlusion. For example, in some embodiments, instead of CT data, magnetic resonance imaging (MRI) data is acquired at stage 30. The MRI data may include, for example, a T2* gradient recalled echo (GRE) magnetic resonance study. In such embodiments, the high-attenuation vessel detection process may include processing the MRI data to, for example, identify whether a susceptibility vessel sign (SVS) is present in the MRI data.

[0035] It is known that the presence of certain imaging features such as HAS and SVS in non-contrast studies indicates the presence of obstruction with high specificity but low sensitivity.

[0036] In this embodiment, the feature detection circuitry 23 determines that a region of high absorption blood vessels is present if a HAS is detected in the NCCT data. The feature detection circuitry 23 may estimate the location of the region of high absorption blood vessels.

[0037] At stage 34, in response to detecting an area of ​​high absorption vessel, the notification circuitry 25 designates the patient as a potential thrombectomy candidate. Designating the patient as a potential thrombectomy candidate may also be described as flagging the patient as a potential LVO patient or flagging the patient as a potential thrombectomy patient. The notification circuitry 25 may also add the patient to a worklist. The notification circuitry 25 may also issue a mobile notification of the LVO candidate, as described below with reference to FIG. 3.

[0038] The rendering circuitry 26 renders a rendered image 36 showing a slice from the NCCT data aligned with the anterior circulation. The rendered image 36 is aligned to the slice of the NCCT data that is most likely to contain an occluding imaging feature.

[0039] Display circuitry 28 displays rendered slices 36 on display screen 16 or on any suitable display. In some embodiments, detected regions of high-absorption vessels are highlighted on rendered image 36. For example, detected regions of high-absorption vessels may be shown in a different color than the rest of rendered image 36. Alternatively, detected regions of high-absorption vessels may be given an outline in rendered image 36.

[0040] The rendered image 36 is available for review by a clinician, as described below with reference to stage 60 .

[0041] If feature detection circuitry 23 does not detect regions of high-absorption vessels in stage 32, then in some embodiments, stages 34 and / or 36 may be omitted. In other embodiments, slices aligned with the anterior circulation are rendered and displayed even if high-absorption vessels are not detected. In some embodiments, an indication may be displayed to the user that high-absorption vessels were not detected.

[0042] At stage 38, the gaze detection circuit 24 performs gaze detection processing. The gaze detection processing includes processing the NCCT scan data to obtain an estimate of the gaze direction of at least one eye of the patient. In this embodiment, gaze detection is always performed regardless of the results of the high-absorption blood vessel detection processing. In other embodiments, gaze detection may be omitted in situations such as when high-absorption blood vessels have already been detected. Gaze detection may be performed after the high-absorption blood vessel detection processing of stage 32, before the high-absorption blood vessel detection processing, or simultaneously with the high-absorption blood vessel detection processing.

[0043] In the following, references to eyes refer to the eyeball, also called the eyeball. If there are two eyes, it can be assumed that both eyes are looking in the same direction.

[0044] The output of the gaze detection process includes a classification of the gaze direction in the NCCT data into one of three classes. In the first class, the gaze is to the right. In the second class, the gaze is to the left. In the third class, the gaze is neither to the left nor to the right, or the gaze is unknown. In what follows, the classifications will be referred to as right, left, and neither / unknown.

[0045] In clinical practice, clinical gaze deviation of the eyes is a well-documented symptom of stroke. Clinical gaze deviation of the eyes is called Prevost's sign. Gaze deviation is defined as an equal, persistent deviation of both eyes from their midline position toward the same side. When present, the eyes deviate toward the side of the brain hemisphere damaged by the stroke. The damaged hemisphere is the side of the body opposite to the side where symptoms such as paralysis or facial ptosis are present.

[0046] The gaze angle of the eye is defined, for example, relative to the contralateral plane of the patient's skull. The gaze angle can be determined, for example, as described in Kobayashi, M., Horizontal gaze deviation on computed tomography: the visual criterion and lesion characteristics in ischemic stroke. Acta Neurol Belg (2018) 118: 581. https: / / doi.org / 10.1007 / s13760-018-0949-1 or Spokoyny, Ilana et al., Visual Determination of Conjugate Eye Deviation on Computed Tomography Scan Predicts Diagnosis of Stroke Code Patients, Journal of Stroke and Cerebrovascular Diseases, Volume 25, Issue 12, 2809-2813.

[0047] In some embodiments, the gaze detection circuitry 24 can return the gaze angle, for example as a numerical value. The gaze angle may be returned in addition to, or instead of, determining whether the gaze direction is left, right, or neither / unknown.

[0048] In this embodiment, a trained model is used to determine whether the eyes are tilted to the right, left, or neither / unknown. The trained model may be, for example, a deep learning classifier. For example, an R-CNN (Regions with Convolutional Neural Network features) method similar to that described in R. Girshick, J. Donahue, T. Darrell, and J. Malik, "Rich feature hierarchies for accurate object detection and semantic segmentation," The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2014, may be used. In other embodiments, any suitable method may be used to determine whether the gaze direction is left, right, or neither / unknown. For example, any suitable image analysis method may be used to perform any suitable classification of gaze direction. Relevant portions of the patient's anatomy may be identified and displayed to the user.

[0049] In some circumstances, a brain scan may be obtained that excludes the patient's eyes from the anatomical region scanned using a CT scanner. In at least such cases, gaze can be derived based on anatomy other than the eyes. For example, compression of extraocular muscles may indicate gaze direction.

[0050] The output of stage 38 is a decision as to whether the gaze direction is left, right, or neither / unknown.

[0051] At stage 40 of FIG. 2, the CT scanner 14 acquires a contrast-enhanced CT scan (CTA scan) of the patient's brain.

[0052] The volumetric data from the CTA scan is provided to data store 20, and from data store 20 to gaze detection circuitry 24. In other embodiments, gaze detection circuitry 24 may obtain the set of volumetric CTA data from any suitable data store. The CTA data may be obtained by gaze detection circuitry 24 at any suitable time after the CTA scan is performed.

[0053] At stage 42, the gaze detection circuitry 24 performs gaze detection processing on the CTA scan data. The gaze detection processing involves processing the CTA scan data to obtain an estimate of the gaze direction of at least one eye of the patient. The gaze detection circuitry 24 outputs a classification of the gaze direction as right, left, or neither / unknown.

[0054] In other embodiments, the gaze detection circuitry 24 may return the gaze angle, for example as a numerical value, in addition to or instead of determining whether the gaze direction is left, right, or neither / unknown.

[0055] In this embodiment, a trained model is used to determine whether the eye leans to the right, left, or neither / unknown. In some embodiments, the trained model may be different from the trained model used in stage 38, given that the determination is performed on contrast data instead of non-contrast data. In other embodiments, the same trained model may be used in stage 38 and stage 42.

[0056] The output of stage 42 is a decision as to whether the gaze direction of the CTA data is left, right, or neither / unknown.

[0057] At stage 44, gaze detection circuitry 24 compares the gaze direction determined at stage 38 with the gaze direction determined at stage 42. If the gaze direction determined at stage 42 is the same as the gaze direction determined at stage 38, gaze detection circuitry 24 determines that a persistent deviation in the gaze direction has occurred. In response to determining that a persistent deviation in the gaze direction has occurred, the method of FIG. 2 proceeds to stage 46 and stage 50. If it is determined that no persistent deviation in the gaze direction has occurred, stage 50 may be omitted. In some embodiments, some or all of stage 46 may be omitted.

[0058] In stage 46, notification circuitry 25 designates the patient as a potential thrombectomy candidate based on the same line of sight between the CTA data and the NCCT data. If the patient was previously designated as a potential thrombectomy candidate in stage 34, notification circuitry 25 does not change the patient's designation as a potential thrombectomy candidate.

[0059] Persistent eye deviation across NCCT and CTA has shown good sensitivity and specificity for large vessel occlusion (Attenhofer et al, The Sustained DeyeCOM Sign As a Predictor of Large Vessel Occlusions and Stroke Mimics, J. Stroke Cerebrovasc Dis. 2018 June; 27(6); 1466-1470). Persistent eye deviation can be used as an indicator of a patient's potential thrombectomy candidate, as shown in Figure 2.

[0060] The rendering circuitry 26 receives an indication of the eye region within the volumetric NCCT data and / or CTA data from the gaze detection circuitry 24. For example, the eye region may be indicated by a bounding box obtained from the segmentation. The rendering circuitry 26 renders at least one image 48 from the volumetric NCCT data and / or CTA data that indicates the eye region. In this embodiment, the rendered image 48 represents a portion of an axial slice of the head through the lens of the eye.

[0061] Display circuitry 28 displays at least one rendered image 48 of the eye region on display screen 16. The at least one rendered image 48 may provide a quick view of the line of sight. The at least one rendered image 48 may provide a snapshot of the eye, including to indicate the suspected side of the occlusion.

[0062] By displaying a snapshot of the eye, the clinician can be guided to the hemisphere of the brain that corresponds to the gaze direction. The clinician may also use left-brain / right-brain cues to identify occlusions.

[0063] In other embodiments, derived measurements related to gaze deviation may also be displayed, for example, the deviation angle may be displayed, and in further embodiments, any other clinical information may be displayed.

[0064] Various methods of displaying line of sight views are further described below with reference to Figures 8(a)-10(c). Derived measurements or other clinical information may also be displayed with any line of sight views described below.

[0065] In stage 50, rendering circuitry 26 renders a set of CTA views 52, 54, 56, 58. Display circuitry 28 displays the CTA views 52, 54, 56, 58 on the display screen 16 according to display rules, such as a hanging protocol. The display of the CTA views may be optimized for human recognition of occlusions.

[0066] The CTA views 52, 54, 56, and 58 are stroke views based on landmarks within the CTA. The stroke views can be views that review specific anatomical and vascular regions of the brain related to stroke. The first view 56 is an anterior circulation view, the second and third views 54 and 58 are two posterior circulation views, and the fourth view 52 is a collateral circulation view. The CTA views 52, 54, 56, and 58 are displayed according to selected display parameters. The images of the CTA views 52, 54, 56, and 58 are aligned to the anatomy. The images are slabbed at a selected thickness with a window level set to a selected value. The selected values ​​for thickness and window level can be deemed optimal for the anatomy and / or lesion being viewed.

[0067] In this embodiment, the CTA views 52, 54, 56, and 58 are only displayed if persistent ocular deviation is detected. If persistent ocular deviation is detected, a first display mode is used in which the CTA views 52, 54, 56, and 58 are displayed. If persistent ocular deviation is not detected, a second display mode is used in which the CTA views 52, 54, 56, and 58 are not displayed. The display circuitry 28 is configured to display the scan that is most appropriate for a given result or criterion.

[0068] In other embodiments, as described below with reference to, for example, Figures 6 and 7, the display circuitry 28 displays both the rendered image 36 of the anteriorly circularly aligned slices and the CTA views 52, 54, 56, and 58, but in different display types depending on the determined gaze deviation. If a persistent gaze deviation is detected, a first display mode is used in which the CTA views 52, 54, 56, and 58 are presented as a large main image display and the rendered image 36 including the anteriorly circularly aligned slices is presented as a smaller image, e.g., a thumbnail image. If a persistent gaze deviation is not detected, a second display mode is used in which the rendered image 36 including the anteriorly circularly aligned slices is presented as a large main image display and the CTA views 52, 54, 56, and 58 are presented as a smaller image, e.g., a thumbnail image. To view the smaller image in more detail, the clinician may select it, for example, by clicking on it.

[0069] Display aspects may be used based on the time the scans were taken or the time intervals at which the scans were taken, for example using the elapsed time between scans.

[0070] If a high-attenuation vascular signature is detected in stage 32, the clinician reviews rendered images 36, including slices aligned with the anterior circulation, in stage 60. If persistent ocular deviation is detected in stage 44, the clinician reviews CTA views 52, 54, 56, and 58. Images 36, 52, 54, 56, and 58 may be the most relevant images presented to the clinician in the initial review. The clinician considers the patient for thrombectomy. For example, the clinician identifies the patient as suitable for thrombectomy. The clinician, for example, elects to transfer the patient.

[0071] Pre-established criteria for thrombectomy and / or transfer may be used. A panel 62 of thrombectomy and / or transfer criteria may be displayed to the clinician. The thrombectomy and / or transfer criteria and their display on panel 62 are described below with reference to Figures 4-7.

[0072] Figure 3 illustrates a notification process 70 that may occur as part of stage 36 or stage 46 of Figure 2. In this embodiment, the notification process 70 includes a worklist or mobile notification of LVO candidates.

[0073] The mobile notification comprises a message sent to a mobile device, such as, for example, a smartphone 72. The message indicates that an urgent review of a potential thrombectomy candidate is required.

[0074] The workflow notification includes an indicator that is on the workflow 74. The workflow includes a list of patients whose data is to be reviewed. An indicator is added to one of the patients, suggesting an urgent review for that patient. In some embodiments, the workflow may be reordered so that the patient identified for urgent review is moved up the list of patients.

[0075] The mobile or workflow notification may include summary information about the patient's status.

[0076] Upon receiving a notification, such as a mobile or workflow notification, a clinician may decide to prioritize review of the patient for which the notification was issued, which may reduce the time it takes for a patient to be reviewed or may reduce the time between scan acquisition and treatment, such as thrombectomy.

[0077] The methods of Figures 2 and 3 may provide a way to present clinically relevant information for determining the presence of large vessel occlusion and identifying thrombectomy candidates in acute stroke patients. Continuous gaze detection is performed across successive images. In the method of Figure 2, the successive images include NCCT data and CTA data, respectively. The method of Figure 2 also identifies non-contrast imaging features associated with occlusion. Cases are flagged based on the presence of persistent ocular deviation or an occlusion imaging feature, which in this embodiment is a high-absorption vascular signature. If a patient is found to meet the criteria for a potential thrombectomy candidate, the clinician is notified.

[0078] The display manner may be selected to expedite review of potential thrombectomy candidates, so that the clinician can be presented with the most relevant information first, allowing the clinician to use the most relevant information when assessing whether an obstruction is present.

[0079] The relevant information is provided to the clinician without the medical image processing device 10 directly detecting the LVO.

[0080] In some embodiments, additional predefined inclusion or exclusion criteria may be incorporated into the methods described above with reference to Figures 2 and 3. For example, ASPECTS or ICH may be used. Such embodiments are described below with reference to Figures 4-7.

[0081] Figure 4 shows elements of a user interface on which information is displayed to a user, such as a clinician. The user interface may be displayed on any suitable screen, such as display screen 16. Figure 4 also shows a smartphone 72 on which a notification message is displayed.

[0082] The user interface includes a panel 62. Two views 82, 84 of the patient's ocular region are displayed on the panel 62. The first view 82 is obtained by rendering image data from a first scan obtained at a first time point. In the embodiment shown in FIG. 4, the first scan is an NCCT scan. The patient's eye is deviated to the left in the first scan. A plus sign (+) is displayed next to the first view 82 to indicate that the eye is deviated in the first view 82. A plus sign may also be displayed if the eye is deviated to the right. If there is no ocular deviation, a minus sign (-) may be displayed.

[0083] In other embodiments, any suitable indicator or indicators of ocular deviation may be used. Any suitable visual effect may be used to highlight or emphasize the ocular deviation, such as those described below with reference to Figures 8(a)-10(c).

[0084] The second view 84 is obtained by rendering image data from a second scan obtained at a second time point. In the embodiment shown in FIG. 4, the second scan is a CTA scan. The patient's eye is similarly deviated in the second scan as in the first scan. Two plus signs are displayed next to the second view 84 to indicate that the patient's eye is consistently deviated in both scans. In other embodiments, any suitable indicator or indicators may be used to represent persistent deviation. Any suitable visual effect may be used to highlight persistent deviation.

[0085] The first and second views 82, 84 are axial slices aligned with the patient's eye lens to show the affected side in each scan. The first and second views 82, 84 provide snapshots of the eye for two consecutive scans at two consecutive time points and are rendered as two consecutive images. If ocular deviation persists across both images, it is known that the patient likely has LVO. The first and second views 82, 84 can be from any two consecutive scans and images. In the embodiment of FIG. 4, the two consecutive scans are an NCCT and a CTA.

[0086] In other embodiments, one or both of the scans may be MRI scans. In some embodiments, the scans include two MRI sequences visible to the eye. SVS of high-absorption vessels is replaced by GRE.

[0087] In other embodiments, the scans may be a 3D scanogram obtained at a first time point and an NCCT scan obtained at a second time point. When performing a CT scan of a region of a patient's body, it is common to first perform a three-dimensional (3D) scanogram. The 3D scanogram may have a large field of view of the region where the local scan is performed. The 3D scanogram may include a low-resolution scan of a large region of the patient's body, for example, the entire patient's body. The NCCT 3D scanogram, followed by an NCCT scan, may also be used to provide an NCCT-only solution for referencing centers.

[0088] In some embodiments, the first scan is an optical image acquired by an optical camera within the scanner. Eye deviation is determined from the optical image and subsequent scans. In some embodiments, a video camera, such as an AI video camera, is used in the imaging to determine eye deviation, and the eye deviation in the video acquired by the video camera is compared to the eye deviation in subsequent scans.

[0089] The first and second views 82, 84 of the eye region may always be present to indicate the side of the anomaly to the user. The side of the anomaly may be indicated to the user without outputting a result to the user, e.g., without outputting an explicit indication that the patient may have an LVO. In some situations, the requirements for regulatory approval may be different for systems that display a diagnosis than for systems that do not provide a diagnosis. Providing the user with relevant information without providing a diagnosis may be important for regulatory approval. Because the tool avoids Computer Aided Detection (CADe), regulatory approval may be easier.

[0090] In FIG. 4, panel 62 also includes information regarding criteria for factors other than ocular deviation. In the embodiment of FIG. 4, the criteria relate to Intracerebral Hemorrhage (ICH), occlusion, Alberta Stroke Program Early CT (ASPECTS) score, and collaterals. First display element 90 represents the ICH (Intracerebral Hemorrhage) score. Second display element 92 represents occlusion information. Third display element 94 represents the ASPECTS (Alberta Stroke Program Early CT) score. Fourth display element 96 represents collaterals. Together, first through fourth display elements 90, 92, 94, and 96 and views 82 and 84 can be considered to provide a set of clinically relevant information for determining the presence of LVO and thrombectomy candidacy in acute stroke patients.

[0091] Other criteria (ICH, occlusion, aspect, collateral) are considered in combination with the gaze deviation shown in the first and second views 82, 84 to provide a full LVO triage solution. In some embodiments, the hospital can also configure the display to select which information about the criteria to display. The selection of which information about the criteria to display may be based on available imaging, for example, imaging available at the hospital or available to the individual patient.

[0092] In the embodiment of Figure 4, each of the display elements 90, 92, 94, 96 is flagged in green if the result associated with that element meets the thrombectomy or transfer criteria. For example, when the ICH score meets the predetermined thrombectomy or transfer criteria, the display element 90 turns green. Because Figure 4 is in black and white, the color green is not represented in Figure 4.

[0093] In other embodiments, any suitable method may be used to indicate that the results of the display elements 90, 92, 94, 96 meet the thrombectomy or transfer criteria, such as any suitable color, line type, shape, or shading, or any suitable visual or other effect.

[0094] If the results of all factors represented by display elements 90, 92, 94, and 96 meet the criteria for thrombectomy or transfer, notification 100 is displayed in panel 62. In the example of FIG. 4, the text of the notification is "Urgent Review, Thrombectomy Candidate." The display of notification 100 may trigger a clinician to urgently review the patient's imaging to determine if the patient is a thrombectomy candidate.

[0095] Figure 5 shows an example in which the results of either the thrombectomy or transfer criteria are inconsistent or absent. Panel 62 includes first and second views 82, 84 and display elements 90, 92, 94, and 96 similar to those in Figure 4. As in Figure 4, persistent ocular deviation is indicated by a plus sign in first view 82 and two plus signs in second view 84.

[0096] In the example shown in Figure 5, both the ICH result and the occlusion result meet the criteria for thrombectomy or transfer, and display elements 90, 92 are flagged in green (green is not shown in Figure 5). The green display indicates that the included criteria have been met. The included criteria indicate that an occlusion is present at a given location.

[0097] The results obtained from the algorithm to determine the ASPECTS score are a contraindication to thrombectomy or transfer criteria. Display element 94 is flagged in red (the red color is not shown in FIG. 5, but is instead represented by a thick outline).

[0098] In the embodiment of FIG. 5, if the results of any of the display elements 90, 92, 94, 96 are a contraindication to thrombectomy or transfer criteria, that display element may be flagged in red.

[0099] Contraindications may occur even when other criteria are met: for example, an obstruction may be present, but the patient may be excluded from treatment based on other imaging features, such as poor collaterality, or other clinical information.

[0100] In other embodiments, any suitable method may be used to indicate that the results of the display elements 90, 92, 94, 96 are a contraindication to thrombectomy or transfer criteria, such as any suitable color, line type, shape, or shading, or any suitable visual or other effect.

[0101] In the embodiment shown in FIG. 5, suitable scans or results are not available to provide information about the collateral. Display element 96 is grayed out to indicate that results are not available. In other embodiments, display elements 90, 92, 94, and 96 may be grayed out if no suitable results are available for the factor associated with that display element. In other embodiments, any suitable method may be used to indicate that results for the factor associated with display elements 90, 92, 94, and 96 are not available. In some embodiments, the display element may be completely hidden. In other embodiments, any suitable color, line type, shape, or shading, or any suitable visual or other effect, may be used.

[0102] Having a list of criteria reviewed (in this embodiment, ICH, occlusion, aspect, collateral) can serve as a checklist for the clinician to check all relevant information, regardless of whether the criteria have an automated outcome.

[0103] FIG. 6 illustrates a clinician's interaction with panel 62. Panel 62 displays first and second views 82, 84 of the patient's ocular region and display elements 90, 92, 94, and 96 representing ICH score, occlusion, ASPECTS score, and collateral, respectively. Both first and second views 82, 84 show gaze deviation, which is additionally indicated by a plus sign in first view 82 and two plus signs in second view 84. All factors associated with display elements 90, 92, 94, and 96 meet the criteria for thrombectomy or transfer and are highlighted in green (not shown in FIG. 6).

[0104] The clinician reviews the factors represented by the display elements 90, 92, 94, 96 (ICH, occlusion, aspect, collateral) for thrombectomy or transfer criteria. When the clinician selects each display element 90, 92, 94, 96, for example, by clicking on the display element, relevant information is displayed to the clinician. For example, appropriate imaging may be displayed.

[0105] 6 shows an example in which a clinician selects an occlusion, for example, by clicking on display element 92. Shading of display element 92 is used to indicate that display element 92 has been selected. In other embodiments, any suitable visual indication may be used to indicate that one of display elements 90, 92, 94, 96 has been selected.

[0106] When one of the criteria (ICH, occlusion, aspect, collateral) is selected by the clinician, the display circuitry 28 selects the most appropriate scan to view that criteria. In the example shown in FIG. 6, the patient has persistent ocular deviation but no high-attenuation vessels. When the clinician selects occlusion by selecting display element 92, the display circuitry 28 displays a set of stroke views 52, 54, 56, 58 for immediate review of key LVO locations within the vasculature.

[0107] The display circuitry also highlights the eye view corresponding to the scan being viewed. The scan being displayed is highlighted so that the eye and left / right side information are always highlighted.

[0108] In Figure 6, a stroke view is obtained from a second (CTA) scan, highlighting a second view of the eye 84. In the example of Figure 6, the second view 84 is highlighted using a surrounding halo effect. In other embodiments, any suitable visual or other effect may be used to highlight the scan being viewed.

[0109] In addition to the stroke views 52, 54, 56, and 58 presented to the clinician, all other scans are available for normal viewing. For example, slice 36, which is aligned to the anterior circulation, is displayed in Figure 6 as a small thumbnail image selectable by the clinician.

[0110] When a clinician begins reviewing criteria and selects a display element associated with any one or more criteria, display circuitry 28 selects the most appropriate scan to view that criterion. For example, if ICH is selected, display circuitry 28 automatically displays the NCCT or, in this embodiment, the GRE, which is obtained by an MRI scan. The most appropriate scan to display may depend on the workflow and imaging used. The selection of the most appropriate scan to display may be configured by the hospital.

[0111] 7 shows that in the case where the patient does not display persistent ocular deviation but does have a high-attenuation vascular sign, the clinician selects occlusion from panel 62. Panel 62 displays first and second views 82, 84 of the patient's eye and display elements 90, 92, 94, and 96 representing thrombectomy or transfer criteria. All criteria in display elements 90, 92, 94, and 96 meet the thrombectomy or transfer criteria and are highlighted in green (not shown in FIG. 7).

[0112] In a first view 82 of the patient's eye, there is deviation, as indicated by the plus sign next to first view 82. In a second view 84 of the patient's eye, there is no deviation. Second view 84 includes a minus sign next to it. The gaze detection circuit 24 determines that there is no persistent gaze deviation.

[0113] When the clinician selects display element 92 representing an occlusion, display circuitry 28 selects the most appropriate scan. Because the patient does not have persistent gaze deviation but does have a high-attenuation vascular signature, the scan selected as the most appropriate scan to display first is the NCCT. To indicate that the NCCT scan is being viewed, first view 82 is highlighted. In FIG. 7, first view 82 is highlighted by a halo effect around first view 82.

[0114] In this embodiment, the NCCT scan is shown by displaying a preset slice output 36 that is aligned with the slice most likely to show the anterior circulation. The slice 36 is selected as the slice most likely to identify high-absorption vessels. By the rendering circuitry 26 selecting the most likely slice 36, the amount of scrolling required between images may be reduced so that minimal scrolling is required.

[0115] In some embodiments, the display of slice 36 may include segmentation of high-absorption vessels. In some embodiments, the display of slice 36 may include labeling of high-absorption vessels. The presence of segmentation and / or labeling may depend on the regulatory approach in some circumstances.

[0116] In the embodiment of FIG. 7, stroke view presets 52, 54, 56, and 58 remain available after viewing the high-absorption vessels. The clinician can also view one or more of stroke views 52, 54, 56, and 58 by selecting from thumbnail views or by any suitable method. The most appropriate stroke view preset may be selected based on the location of the high-absorption vessels. In some embodiments, display circuitry 28 automatically selects the most appropriate stroke view preset based on predetermined rules. In some embodiments, the stroke view preset may be determined by the clinician.

[0117] In the embodiment shown in Figures 6 and 7, the display circuitry 28 selects the most appropriate scan and view for the selected criteria. When an occlusion is selected and there is a high-attenuation vessel without persistent gaze deviation, the display circuitry 28 uses alignment to the slice most likely to contain the circulation in the non-contrast scan for immediate review of imaging features related to the occlusion. When an occlusion is selected and there is persistent gaze deviation, the display circuitry 28 displays a presentation of the stroke view in the CTA for immediate review of the vasculature. If neither a high-attenuation vessel nor persistent gaze deviation is detected, the initial view of the NCCT scan is displayed. The initial view of the NCCT scan is, for example, the initial display of the NCCT scan when loaded into the viewer. In other embodiments, any suitable view or views may be displayed.

[0118] Different scans may be displayed for different criteria. Different scans may be displayed for different results. Display circuitry 28 also selects the most appropriate view for each scan (e.g., a stroke view).

[0119] Panel 62 provides an interactive display of information related to thrombectomy or transfer criteria. In some embodiments, a summary view may be provided in addition to or instead of panel 62. The presentation of information may be determined and presented to the user in summary form. For example, the information may be displayed as a report or another non-interactive view format.

[0120] Figures 8(a)-8(c) and 9(a)-10(c) illustrate methods of displaying gaze on a user interface to highlight persistent deviations across scans. Highlighting persistent deviations can enable a user, such as a clinician, to quickly determine the persistent deviations and quickly assess whether a patient is a candidate for thrombectomy or transfer. The eye displays according to Figures 8(a)-8(c) and 9(a)-10(c) may be substituted for views 82, 84 shown in Figures 4-7.

[0121] 8(a)-8(c) depict the user interface in an embodiment in which hatching is used to highlight the direction of deviation.

[0122] In each of Figures 8(a)-8(c), the gaze detection circuit 24 has segmented the eyeballs 112, 114 and lenses 116, 118 of each eye. The gaze detection circuit 24 may use any suitable segmentation. The rendering circuit 26 renders the image such that each of the segmented eyeballs 112, 114 is outlined. Each of the segmented lenses 116, 118 is represented by a graphic element, which is a white element in the images of Figures 8(a)-8(c).

[0123] Ocular deviation is indicated by hatching of the segmented eye. In Figure 8(a), the patient's ocular deviation to the right is indicated by hatching with diagonal lines aligned with the gaze direction. In Figure 8(b), there is no deviation and no hatching is used. In Figure 8(c), the patient's ocular deviation to the left is indicated by hatching with diagonal lines aligned with the gaze direction.

[0124] In other embodiments, any suitable graphic element may be used to emphasize the eyeball and / or lens. Any suitable pattern or visual feature may be used to highlight the gaze direction. For example, an arrow may be used to indicate the gaze direction.

[0125] In some embodiments, different colors are used to represent different scans, such that the color of rendering the eyes varies for different scans across different time points. A first color may be used when rendering eyes in images derived from a scan at time point 1. A second color may be used when rendering eyes in images derived from a scan at time point 2. A third color may be used when rendering eyes in images derived from a scan at time point 3. The use of color may be combined with the use of other visual effects, such as hatching.

[0126] Figures 9(a)-10(c) show the user interface in an embodiment where both hatching and color are used to highlight the direction of deviation. Figures 9(a)-10(c) are black and white, without showing color.

[0127] A single image 120 is used to represent the patient's ocular region, and two or more scans (e.g., NCCT and CTA scans) obtained at different time points are registered together so that they can be displayed on the same image.

[0128] Slider bar 130 represents time and is used to switch between views of two or more scans. The user moves indicator 132 on slider bar 130 to change between views representing different times. In the embodiment of FIGS. 9(a)-9(c), the scale on slider bar 130 represents the elapsed time from the first scan, represented by time=0. In other embodiments, the scale on slider bar 130 may represent the actual time of the scan.

[0129] 9(a) and 9(b) show an example with persistent deviation (DeyeCOM + / +). In FIG. 9(a), a baseline scan is shown. The baseline scan represents the eye deviation at time=0, as indicated by the slider bar 130. Eye detection and lens segmentation are performed and used to highlight the patient's eyes in the baseline scan. In the embodiment of FIGS. 9(a)-9(c), each eyeball 122, 124 is outlined, and each lens 126, 128 is outlined. In other embodiments, any suitable method of highlighting the eyeball and / or lens may be used.

[0130] In Figure 9(a), the direction of eye deviation in the baseline scan is indicated by red and white hatching aligned with the gaze direction (the red is not visible in Figure 9(a)). A red plus sign appears on the slider bar 130 at time=0. The slider bar indicator 132 is positioned at time=0.

[0131] In FIG. 9(b), indicator 132 moves to the time=+2 minute position, and image 120 shows a second scan obtained two minutes after the baseline scan. Eye detection and segmentation in the second scan are registered to the baseline scan. In the second scan, the direction of eye deviation in the second scan is indicated by green and white hatching aligned with the gaze direction (the green is not visible in FIG. 9(b)). A green plus sign also appears on slider bar 130 at time=+2 minutes. Eyes 122, 124 and lenses 125, 128 are outlined.

[0132] 9(a) and 9(b), moving the indicator 132 of the slider bar 130 between time points 0 and +2 minutes causes the color of the image 120 to fade between red at 0 and green at +2 minutes. The displayed image may fade between the image shown at 0 and the image shown at +2 minutes.

[0133] The user can see at a glance whether the deviation is persistent. The user can move the indicator 132 back and forth on the slider bar 130 to switch between the baseline scan and the second scan. The use of color allows the user to easily distinguish between the baseline scan and the second scan. The color may gradually fade between time points. The slider bar 130 and changing colors may provide an intuitive way to indicate gaze deviation between the baseline scan and the second scan.

[0134] In other embodiments, there is no fading between images. In some embodiments, the image at 0 and the image at +2 minutes may be overlaid. In further embodiments, any suitable visual method may be used to combine or transition the images at 0 and +2 minutes.

[0135] In some situations, the baseline scan and second scan may be followed by a follow-up scan taken at a longer interval after the second scan, which may not be acute but may be used to confirm that an acute condition has resolved.

[0136] FIG. 9(c) shows an example of a follow-up scan. The follow-up scan is registered along with the baseline scan. The results of the follow-up scan are shown in image 120 as the user slides indicator 132 of slider bar 130 to the time of the follow-up scan. In FIG. 9(c), the time of the follow-up scan is +23 hours, 23 hours after the baseline scan. Slider bar 130 is not presented as a linear scale. Instead, the scale of slider bar 130 is selected so that scan times 0, +2 minutes, and +23 hours are easily distinguishable and fit on the same scale.

[0137] The follow-up scan in Figure 9(c) shows no ocular deviation. The eye is colored blue (blue is not shown in Figure 9(c)). It is not hatched. A minus sign appears with the time +23 hours on the slider bar 130 (blue is not shown in Figure 9(c)).

[0138] By sliding indicator 132 along slider bar 130, the user can switch between two or more scans to see eye deviation over multiple time points. The image may fade between different scans.

[0139] FIGS. 10(a) and 10(b) represent the same views as FIGS. 9(a) and 9(b), except for an example without persistent deviation (DeyeCOM + / -). FIG. 10(a) shows an image 120 at time=0. FIG. 10(a) represents a baseline scan, such as an NCCT scan, and is the same as FIG. 9(a). The indicator 132 of the slider bar 130 is positioned at time=0. Ocular deviation is indicated by red and white hatching of the eyes 122, 124 (red is not shown in FIG. 10(a)). The hatching is aligned with the gaze direction. Ocular deviation is also indicated by a plus sign at time=0.

[0140] In Figure 10(b), a second scan, such as a CTA scan, at time = +2 minutes is shown in image 120. The second scan is registered to the baseline in Figure 10(a). Indicator 132 of slider bar 130 is positioned at time = +2 minutes.

[0141] In Figure 10(b), there is no gaze deviation. The eyeball is shaded in green, which represents the point in time (green is not shown in Figure 10(b)). A grayed-out version of the hatching is shown on the eyeball. The grayed-out hatching indicates to the user that there was gaze deviation previously, but it is no longer present. In other embodiments, any suitable method of visually indicating gaze direction or change in gaze direction may be used.

[0142] A minus sign is shown on the slider bar 130 at Time=+2 minutes.

[0143] As the user moves indicator 132 on slider bar 130 between time=0 and time=+2 minutes, the color fades between the colored scans to show the transition in gaze direction between the red and white hatched image at time=0 and the green and gray hatched image at time=+2 minutes. The user interface allows for color fading between the colored scans to confirm changes in gaze direction.

[0144] FIG. 10(c) shows the results of a follow-up scan in image 120. The follow-up scan is registered to the baseline in FIG. 10(a). The follow-up scan is taken at time +23 hours. The follow-up scan shows no gaze deviation. The eye is shaded in blue (blue is not shown in FIG. 10(c)), and a blue minus sign is shown on slider bar 130 at time = +23 hours.

[0145] 9(a)-10(c), the data is registered and the eye and lens are segmented to give each scan a distinct color. As the indicator 132 of the slider bar 130 moves between time points, the image 120 fades between images from different scans.

[0146] 9(a)-9(c) use scans at specific times (0, +2 minutes, +23 hours) as examples, but in other embodiments, scans may be obtained at any suitable time points. Any suitable time intervals and time ranges may be displayed.

[0147] In some embodiments, the geometry of the eye is morphed across scans to exhibit movement. For example, a representation of the lens may move from a first position in a baseline scan to a second position in a second scan. The representation of the lens may move in a continuous manner so that the eye appears to rotate between the first and second positions.

[0148] In some embodiments, a sequence between images of two or more scans automatically cycles between views of the scans when a user performs a triggering action. For example, the triggering action can be hovering over image 120 or hovering over slider bar 130. The images of two or more scans can automatically cycle back and forth. The automatic cycling can include fading between images. The automatic cycling can include morphing between images. In some embodiments, a movie of the fading or morphing sequence can be automatically saved as a capture. The movie can be saved to data store 20 or any other suitable data store. The movie can be saved to a PACS.

[0149] 9(a)-10(c), the segmented lens and the segmented eye are each outlined using solid lines in image 120. In other embodiments, the lens outline and / or eye outline may be visualized with different line styles for each scan. For example, the lens and eye in the baseline scan may be outlined using solid lines, and the lens and eye in the second scan may be outlined using dotted lines.

[0150] Methods similar to those described above with reference to FIGS. 2 and 3 and interfaces similar to those described above with reference to FIGS. 4-10(c) may be used to process and display any suitable data. For example, any suitable modality or modalities of data may be used, such as CT data, cone-beam CT data, X-ray data, ultrasound data, MR data, PET data, or SPECT data. Any suitable first and second scans may be used. The scans may be of any suitable patient or other subject. The displayed images and data may be reviewed by any suitable user, for example, any suitable clinician or researcher. In some embodiments, rather than displaying images or representations of both of a patient's eyes, an image or representation of one of the patient's eyes may be displayed.

[0151] One embodiment provides a medical imaging device comprising a method for presenting clinically relevant information that flags potential LVO / thrombectomy patients, at least part of which information includes (persistent) gaze deviation.

[0152] Such eye deviation may be detected automatically using image analysis techniques in which relevant parts of the anatomy are also identified and displayed to the user.

[0153] High absorption vessels may be detected automatically using image analysis methods, and relevant portions of the anatomy may also be identified and displayed to the user.

[0154] A CTA view optimized for human occlusion recognition may be automatically determined and presented to the user.

[0155] Other clinically relevant information may also be detected / displayed.

[0156] The presentation may be determined and presented to the user in a summary format (such as a report or other non-interactive view format).

[0157] The presentation enables the UI.

[0158] A notification may be sent to the user (eg, on a smartphone, layout and summary information).

[0159] The worklist is prioritized according to automatically detected information.

[0160] Direct LVO detection or other CADe may also be included in the display.

[0161] Derived measurements regarding gaze deviation or any other clinical information may be displayed.

[0162] The data may be registered and the eye and lens may be segmented and given a distinct color for each scan.

[0163] The user interface allows fading between colored scans to show transitions in gaze direction.

[0164] The eye geometry may be morphed across the scan to show movement instead of grayscale transitions.

[0165] When the user hovers over the view, the sequence may automatically cycle back and forth.

[0166] A video of the fading / morphing sequence may be automatically saved as a capture to the PACS.

[0167] In the report, the lens and eye outline from both scans may be visualized with a different line style for each scan (e.g., solid line for one and dotted line for the other).

[0168] One embodiment provides an image display device (medical image display device) including a processing circuit configured to receive medical image data including at least a patient's eye and to control a display manner of the medical image data based on a deviation angle of the patient's eye.

[0169] The processing circuitry may be further configured to receive a plurality of medical image data scanned at at least two different times, and control a display manner of the plurality of medical image data based on the times.

[0170] The processing circuitry may be further configured to receive a plurality of medical image data scanned at at least two different points in time, and control the display manner of the plurality of medical image data based on the interval between the points in time.

[0171] An embodiment provides a medical data processing device comprising a processing circuit configured to: receive first medical imaging data representing a first scan of a subject, automatically process the first medical imaging data to determine a first gaze direction of the subject, receive second medical imaging data representing a subsequent second scan of the subject, automatically process the second medical imaging data to determine a second gaze direction of the subject, use the first gaze direction and the second gaze direction to determine whether a persistent gaze deviation has occurred, identify the subject as a potential large vessel occlusion (LVO) or a candidate for thrombectomy, the identification depending on whether a persistent gaze deviation has occurred, and if the subject is identified as a potential LVO or thrombectomy candidate, notify a user that the subject is a potential LVO or thrombectomy candidate.

[0172] The processing circuitry may be further configured to display a first image of the subject's ocular region rendered from the first medical imaging data and / or a second image of the subject's ocular region rendered from the second medical imaging data.

[0173] The processing circuitry may be further configured to process the first medical imaging data and / or the second medical imaging data to obtain at least one image feature, and identifying the subject as a potential LVO or a candidate for thrombectomy may further depend on the at least one image feature.

[0174] The at least one image feature may include at least one of a non-contrast image feature, a high absorption vessel, a hyperdense artery sign (HAS), and a susceptibility vessel sign (SVS).

[0175] The processing circuitry may be further configured to display at least one image showing an anatomical region including the at least one image feature, the at least one image being rendered from the first medical imaging data and / or the second medical imaging data.

[0176] The processing circuitry may be further configured to select and display at least one image for user review, the at least one image being selected to enable a human to recognize the occlusion.

[0177] The at least one image for review may include multiple Computed Tomography Angiography (CTA) views.

[0178] The at least one image for review may include multiple stroke views.

[0179] The processing circuitry may be further configured to display a number of criteria for LVO / thrombus removal to the user.

[0180] The processing circuitry may be further configured to automatically determine whether the subject meets at least some of the plurality of criteria and display an indication to a user whether the subject meets each criterion.

[0181] The processing circuitry may be further configured to prioritize workflows depending on the identification of persistent gaze deviation and / or depending on other information obtained from processing the first medical imaging data and / or the second medical imaging data.

[0182] The notifying may include sending the notification to a mobile device, such as a smartphone.

[0183] The first scan may be a non-contrast CT (NCCT) scan and the second scan may be a contrast CT scan. The first scan may be an NCCT scan and the second scan may be an NCCT scan. The first scan may be an MRI scan and the second scan may be an MRI scan. The first scan may be an optical imaging procedure and the second scan may be a CT scan or an MRI scan. The first scan may be a video imaging procedure and the second scan may be a CT scan or an MRI scan.

[0184] The processing circuitry may be further configured to determine to automatically process the first medical imaging data and / or the second medical imaging data to detect an LVO, and identifying the subject as a potential LVO or a candidate for thrombectomy may depend on detecting the LVO.

[0185] Although particular circuits are described herein, in alternative embodiments, the functionality of one or more of these circuits may be provided by a single processing resource or other component, or the functionality provided by a single circuit may be provided by a combination of two or more processing resources or other components. A reference to a single circuit encompasses multiple components that provide the functionality of that circuit, whether or not such components are separate from one another. A reference to multiple circuits encompasses a single component that provides the functionality of those circuits.

[0186] According to at least one of the embodiments described above, it is possible to improve the visibility of the line of sight direction in a medical image.

[0187] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are intended to be included within the scope and spirit of the invention, as well as within the scope of the inventions and their equivalents as defined in the claims. [Explanation of symbols]

[0188] 10 Medical image processing device 12 Computing Devices 14 CT scanner 16 display screens 18 Input Devices 20 Data storage unit 22 CPU 23 Feature detection circuit 24 Eye gaze detection circuit 25 Notification circuit 26 Rendering Circuit 28 Display circuit

Claims

1. an acquisition unit that acquires two medical images obtained by two consecutive scans, a non-contrast scan and an angiography scan, each of which includes at least one eyeball of a subject; a gaze detection unit that detects the gaze direction of the eyeball included in each of the two medical images; a display control unit that determines a display mode of the medical image of the subject based on at least one of the two consecutive scans based on the gaze directions of the two eyeballs detected from the two medical images, and displays the medical image of the subject on a display unit in the determined display mode; Equipped with the gaze detection unit determines whether or not there is a deviation to the left or right in the gaze direction of the eyeball for each of the two medical images, and determines that there is a persistent deviation in the gaze direction of the eyeball when the determined gaze deviations match between the two medical images; The display control unit controls the display mode for displaying both the medical image based on the non-contrast scan and the medical image based on the angiography scan to: If it is determined that there is a persistent deviation in the gaze direction of the eyeball, a first display mode is determined in which the medical image based on the angiography scan is displayed as a larger main image than the medical image based on the non-contrast scan; If it is determined that there is no persistent deviation in the gaze direction of the eyeball, a second display mode is determined in which the medical image based on the non-contrast scan is displayed as a larger main image than the medical image based on the angiography scan. Medical image display device.

2. the display control unit determines a display mode of the medical image related to the subject based on a time point when either of the two medical images was acquired or an interval between times when the two medical images were acquired. The medical image display device according to claim 1 .

3. a feature detection unit that detects at least one imaging feature included in the medical image obtained by the non-contrast scan; the display control unit determines a display manner of the medical image related to the subject further based on the detected at least one imaging feature.

3. The medical image display device according to claim 1.

4. The medical image display device of claim 3 , wherein the at least one imaging feature includes at least one of a hyperdense artery sign and a susceptibility vessel sign.

5. the feature detection unit detects the at least one imaging feature from the medical image obtained by the non-contrast scan; 5. The medical image display device according to claim 3.

6. 6. The medical image display device according to claim 3, wherein the display control unit determines a display mode of the medical image relating to the subject that displays an anatomical region in which the at least one imaging feature is detected.

7. The medical image display device according to claim 3 , further comprising a notification unit that issues a notification to an external device depending on the gaze direction of the eyeball or the detected at least one imaging feature.

8. The medical image display device according to claim 1 , further comprising a notification unit that detects a large blood vessel occlusion based on the medical image obtained by the non-contrast scan.

9. an image processing unit that renders an image showing a region including the eyeball based on the medical image obtained by the non-contrast scan; the display control unit displays an image showing a region including the eyeball on the display unit. The medical image display device according to claim 1 .

10. the gaze detection unit segments a region of at least one eyeball included in each of the two medical images; the display control unit displays the segmented image showing at least one of the eyes on the display unit. The medical image display device according to claim 1 .

11. The display control unit displaying the segmented lens of the at least one eye; highlighting the segmented gaze direction of the at least one eye; The medical image display device according to claim 10 , wherein the medical image display device performs at least one of the following:

12. the image processing unit renders an image showing a region including the eyeball based on each of the two medical images; the display control unit displays the region including the eyeball using different visual effects for each of the two different time points when the two medical images were obtained. The medical image display device according to claim 9 .

13. The display control unit using different colors for the segmented eyeball at each of the two different time points as the different visual effects; As the different visual effects, outlining the segmented eyeball with different line styles at each of the two different time points; fading between different visual effects at each of the two different points in time in response to input from a user; and The medical image display device according to claim 12, wherein the medical image display device performs at least one of the following:

14. the image processing unit renders an image showing a region including the eyeball based on each of the two medical images obtained at two different time points; The display control unit determining a display mode of the medical image relating to the subject, in which the geometry of the at least one eyeball morphs between the two different time points; determining, in response to an input from a user, a display mode of the medical image relating to the subject in which the image showing the at least one eyeball is cycled between the two different time points; Rendering and outputting a video of the image showing the at least one eyeball cycling between the two different points in time; and perform at least one of the following: The medical image display device according to claim 9 .

15. acquiring two medical images, each including at least one eye of a subject, obtained by two consecutive scans, a non-contrast scan and an angiographic scan; Detecting the gaze direction of the eyeball included in each of the two medical images; determining a display mode of the medical image of the subject based on at least one of the two consecutive scans based on the gaze directions of the two eyeballs detected from the two medical images, respectively, and displaying the medical image of the subject on a display unit in the determined display mode; Including, Detecting the gaze direction includes determining whether there is a left or right deviation in the gaze direction of the eyeball for each of the two medical images, and determining that there is a persistent deviation in the gaze direction of the eyeball when the determined gaze deviations match between the two medical images; Displaying the medical image of the subject on the display unit includes displaying both the medical image based on the non-contrast scan and the medical image based on the angiography scan in a display mode, If it is determined that there is a persistent deviation in the gaze direction of the eyeball, a first display mode is determined in which the medical image based on the angiography scan is displayed as a larger main image than the medical image based on the non-contrast scan; and when it is determined that there is no persistent deviation in the gaze direction of the eyeball, determining a second display mode in which the medical image based on the non-contrast scan is displayed as a main image larger than the medical image based on the angiography scan. Medical image display method.

16. acquiring two medical images, each including at least one eye of a subject, obtained by two consecutive scans, a non-contrast scan and an angiographic scan; Detecting the gaze direction of the eyeball included in each of the two medical images; determining a display mode of the medical image of the subject based on at least one of the two consecutive scans based on the gaze directions of the two eyeballs detected from each of the two medical images, and displaying the medical image of the subject on a display unit in the determined display mode; A program for causing a computer to execute the above, Displaying the medical image of the subject on the display unit includes displaying both the medical image based on the non-contrast scan and the medical image based on the angiography scan in a display mode, If it is determined that there is a persistent deviation in the gaze direction of the eyeball, a first display mode is determined in which the medical image based on the angiography scan is displayed as a larger main image than the medical image based on the non-contrast scan; and when it is determined that there is no persistent deviation in the gaze direction of the eyeball, determining a second display mode in which the medical image based on the non-contrast scan is displayed as a main image larger than the medical image based on the angiography scan. program.

Images