Disease Characterization From Fused Pathology And Radiology Data

[0001] This application claims the benefit of U.S. Provisional Application 62/099,634 filed Jan. 5, 2015.

[0002] Invasive and in situ or non-invasive components may co-exist within the same pulmonary adenocarcinoma nodule. This may result in a heterogeneous ground glass nodule (GGN) represented in a computed tomography (CT) image. Conventional approaches are unable to distinguish GGNs exhibiting minimal invasion from GGNs exhibiting frank invasion on CT. Conventional approaches rely on histopathology to definitively identify the extent of invasive adenocarcinoma (IA) from in situ or non-invasive adenocarcinoma (AIS) within a GGN.

[0003] Guidelines for managing GGNs detected in CT imagery have been proposed but are not widely implemented. These guidelines consider the size of the GGN and the size of the solid component as represented in the CT imagery. Conventional approaches may examine the relationship between the solid component size and cancer prognosis. However, conventional approaches do not map IA detected by histological examination to data provided by CT imaging.

[0004] Registering ex vivo histology slices to pulmonary CT imagery is challenging. Pulmonary tissue, or other cancerous tissue, is soft and collapses during histology preparation, causing elastic deformations and ripping of the tissue. Correspondences between histology slices and CT imagery are also difficult to identify because histology specimens are frequently obliquely sectioned without consideration of the CT reconstruction. Furthermore, in a clinical setting, only a relatively small number of histology slices may be available from a GGN, and these slices may have been frozen or prepared after fixation, causing further deformities.

[0005] Some conventional approaches have attempted to reconstruct thinly sliced lung adenocarcinoma histology slices into three dimensional (3D) models, but these 3D reconstructions are not registered to CT imagery. At least 25% of nodules identified on baseline CT imagery are GGN nodules. GGNs that persist on follow-up scans often represent early cancers that, if resected, have a five year disease free survival range of between 67% and 100%, depending on the extent of IA in the GGN. Thus, a more accurate way to distinguish IA from AIS in GGNs detected in CT imagery would facilitate improved patient outcomes and reduce un-needed invasive procedures.

[0015] Example methods and apparatus distinguish invasive adenocarcinoma (IA) from in situ or non-invasive adenocarcinoma (AIS) represented in a three dimensional (3D) computed tomography (CT) image of a region of tissue demonstrating cancerous pathology. The 3D CT image may include representations of ground glass nodules (GGN). Example methods and apparatus perform a 3D histological reconstruction of histological slices of a GGN and register the 3D histological reconstruction with a 3D CT image of the GGN, thus combining or “fusing” the 3D histological reconstruction with the 3D CT image. Example methods and apparatus map regions of IA detected in the 3D histology reconstruction to the 3D CT image and extract texture or shape features associated with IA from the 3D CT image. A classifier may be trained to detect computer extracted textural or shape features associated with IA in a 3D CT image.

[0016] Example apparatus and methods use histologic reconstruction and deformable co-registration to map the extent of IA represented in a 3D histopathology reconstruction of a set of histological slices of a GGN onto a 3D CT image of the GGN. The 3D histological reconstruction is combined with the 3D CT imagery using elastic registration techniques. Machine learning techniques may be used to train a classifier to distinguish IA from AIS using the combined 3D histology-3D CT data. The classifier may distinguish IA from AIS by identifying texture or shape features associated with IA in the 3D CT imagery. The classifier may use, for example, support vector machine (SVM) regression or other machine learning techniques. Example methods and apparatus facilitate assessment of the presence of significant invasion. Significant invasion may be defined as invasion greater than five millimeters (mm).

[0017] In one embodiment, 3D CT imagery and histology slices of resected GGNs with significant (n=9) and minimal or no (n=5) invasion are examined. Following high-resolution scanning of histology slides of a GGN, invasive and in situ components are outlined automatically or by an expert pathologist. FIG. 6 is an illustration of a histological slice 600 of a GGN. IA tissue 620 is outlined and distinguished from non-invasive tissue 610. The outlines may be mapped onto CT images of the region of tissue through co-registration of the histology with CT images.

[0018] Example methods and apparatus perform a 3D histological reconstruction of a region of tissue demonstrating IA or AIS using the histology slices. In one embodiment, the 3D histological reconstruction uses a combined natural gradient and group-wise alignment based approach. FIG. 7 is an illustration of an example 3D histological reconstruction 700. A set of histology slices 710 in which IA and AIS tissue have been annotated are registered. IA tissue 720 is extrapolated into the 3D histological reconstruction from the annotation of the individual slices. While the histology slices may be approximately two dimensional, the invasive component extends continuously in three dimensions in the actual GGN. Thus, example methods and apparatus facilitate extending the annotated IA or AIS components from the histology slices into the 3D histology reconstruction.

[0019] Following 3D reconstruction of the histological nodule, the 3D histologic reconstruction is registered to the nodule segmented on CT. FIG. 8 is an illustration of a fused histological-CT image 800. Fused histological-CT image 800 represents a 3D histological reconstruction slice combined with a CT image of the same region of tissue. IA tissue 820 and non-invasive tissue 810 are registered with the CT image 830 of the same region of tissue. In one embodiment, affine and elastic registration facilitates mapping the extent of invasion from ex vivo pathology represented by IA tissue 820 and non-invasive tissue 810 onto corresponding CT imagery represented by CT image 830. Affine and elastic registration facilitates correcting deformations suffered by the tissue during excision and histology preparation. Recovering non-linear deformations between the template and target images facilitates registering the 3D histological reconstruction with the CT imagery.

[0020] In some situations, histology slices may not be taken in the same plane that CT images are taken. Example methods and apparatus may employ an interactive CT nodule sectioning to identify a CT tumor plane that matches the plane of the histology slice. The matching may be performed by visually matching anatomic landmarks that are represented on both the histology slice and the CT image. An anatomic landmark may be, for example, a blood vessel or other structure discernable on both the histology slice and the CT image. The histology slice and the corresponding CT slices are then aligned using elastic registration approaches.

[0021] In one embodiment, a histologic section from an excised lung nodule is registered with a corresponding pre-operative CT image. A region of invasion determined from the histology may then be mapped onto the CT image. The histologic sectioning may be performed in a direction oblique to the CT axial, coronal, or sagittal views. Using anatomic landmarks, including pleura or blood vessels, the cutting plane may be identified and the CT scan may be resampled in that plane. The histology slices may be automatically aligned to their CT oblique correspondent.

[0022] In one embodiment, CT-derived features are extracted to identify and distinguish IA from AIS. CT-derived features may include intensity statistics, Haralick features. Gabor features, or other textural features that facilitate characterizing the textural appearance of the nodule and identifying those features that distinguish IA from in situ disease. The textural features may be extracted based on intensity alone without considering shape, or the textural features may be extracted with consideration of shape. In another embodiment, shape features may also be extracted and used to distinguish invasion from in situ disease.

[0023] Example apparatus and methods perform a more accurate registration of histology and CT than conventional approaches. Example apparatus and methods thus facilitate mapping the invasive and in situ components from the 3D histology reconstruction onto 3D CT. The accuracy of the registration may be assessed by measuring the deviation of anatomic landmarks including, for example, blood vessels, between the histology and CT imagery. Example methods and apparatus achieve a registration accuracy of at least 66%, which is a significant improvement over conventional approaches.

[0024] Upon mapping IA or AIS from the 3D histological reconstruction to the 3D CT imagery, example methods and apparatus may resample the 3D CT imagery to have consistent voxel sizes. In one embodiment, a consistent voxel size of 0.5 mm³ is employed. In another embodiment, other voxel sizes may be employed.

[0025] Example methods and apparatus may also be applied to other types of cancer, including breast cancer. Example apparatus and methods may use images provided by DCE (dynamic contrast enhanced) MRI (magnetic resonance imaging) as well as or in addition to CT imagery. Due to the size of excised breast cancer samples, a lumpectomy specimen may be sectioned into hemispheres instead of the planar histological slices employed with GGNs. In one embodiment, digitized images are digitally stitched to create a pseudo-whole mount histology section using a HistoStitcher. The digital 3D histology lumpectomy specimen is then reconstructed utilizing a group-wise registration method. The 3D reconstruction of the histology facilitates true 3D to 3D registration between the histology and DCE-MRI. An affine transform followed by a selective deformable registration is applied, allowing the mapping of the residual cancer extent from the histology onto MRI.

[0026] By fusing histology with CT and image analytics to train classifiers to distinguish IA from AIS, example methods and apparatus facilitate the early detection of invasive adenocarcinoma on pre-operative CT. Example methods and apparatus thus produce the concrete, real-world technical effect of increasing the accuracy of invasion characterization. Additionally, example apparatus and methods increase the probability that at-risk patients receive timely treatment tailored to the particular pathology they exhibit. Example methods and apparatus may also reduce the number of invasive procedures needed to accurately predict IA in patients. The additional technical effect of reducing the expenditure of resources and time on patients who are less likely to suffer IA or disease progression is also achieved. Example methods and apparatus thus improve on conventional methods in a measurable, clinically significant way.

[0027] Some portions of the detailed descriptions that follow are presented in terms of algorithms and symbolic representations of operations on data bits within a memory. These algorithmic descriptions and representations are used by those skilled in the art to convey the substance of their work to others. An algorithm, here and generally, is conceived to be a sequence of operations that produce a result. The operations may include physical manipulations of physical quantities. Usually, though not necessarily, the physical quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated in a logic, and so on. The physical manipulations create a concrete, tangible, useful, real-world result.

[0028] It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, and so on. It should be borne in mind, however, that these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise, it is appreciated that throughout the description, terms including processing, computing, calculating, determining, and so on, refer to actions and processes of a computer system, logic, processor, or similar electronic device that manipulates and transforms data represented as physical (electronic) quantities.

[0029] Example methods may be better appreciated with reference to flow diagrams. While for purposes of simplicity of explanation, the illustrated methodologies are shown and described as a series of blocks, it is to be appreciated that the methodologies are not limited by the order of the blocks, as some blocks can occur in different orders and/or concurrently with other blocks from that shown and described. Moreover, less than all the illustrated blocks may be required to implement an example methodology. Blocks may be combined or separated into multiple components. Furthermore, additional and/or alternative methodologies can employ additional, not illustrated blocks.

[0030] FIG. 1 illustrates an example computerized method 100 for distinguishing IA from AIS. Method 100 includes, at 110, accessing a CT image of a region of tissue demonstrating cancerous pathology. Accessing the CT image may include accessing a CT image of a region of lung tissue. The CT image may include a texture feature or a shape feature. The region of tissue represented in the CT image may include a GGN. The CT image may be stored, for example, in a computer memory or may be provided across a computer network. In one embodiment, the CT image has an image size of 512 by 512 pixels. The CT image may have an in-plane resolution of 0.57 mm to 0.87 mm. A distance between a first member of a set of CT images and a second, different member of the set of CT images may be from 1 mm to 5 mm. In another embodiment, other images sizes, in-plane resolutions, distances between CT images, or imaging techniques may be employed.

[0031] Method 100 also includes, at 120, extracting a set of texture features from the CT image. The set of texture features may include intensity statistics, Haralick features, Gabor features, or other textural features used to characterize the textural appearance of the GGN represented in the CT image. The set of texture features may be used to identify features that distinguish IA from in situ disease. In another embodiment, shape features may also be extracted and used to distinguish IA from in situ disease.

[0032] Method 100 also includes, at 130, providing the set of texture features to an automated IA classifier. Method 100 also includes, at 140, receiving, from the IA classifier, a classification of the set of texture features. In one embodiment, the IA classifier is trained on a set of composite images. The composite images are formed from a set of 3D histology reconstructions of a region of tissue demonstrating IA or AIS combined with a set of 3D CT images of the region of tissue demonstrating IA or IAS.

[0033] Method 100 also includes, at 150, classifying the region of tissue as IA or AIS. Method 100 may classify the region of tissue based, at least in part, on the set of texture features and the classification. In one embodiment, method 100 may classify the region of tissue based on a shape feature extracted from the CT image.

[0034] Example methods and apparatus facilitate applying a more appropriately determined treatment based on the classification of IA within the area of tissue under investigation. Using a more appropriately determined and applied treatment may lead to less therapeutics being required for a patient or may lead to avoiding or delaying a biopsy, a resection, or other invasive procedure. When regions of cancerous tissue, including the lung captured in CT images, are more quickly and more accurately classified as likely or unlikely to experience IA, patients with poorer prognoses may receive a higher proportion of scarce resources (e.g., therapeutics, physician time and attention, hospital beds). Patients with better prognoses may thus be spared unnecessary treatment, which in turn spares unnecessary expenditures and resource consumption. Example methods and apparatus therefore have the real-world, quantifiable effects of improving patient outcomes and reducing resource expenditure.

[0035] While FIG. 1 illustrates various actions occurring in serial, it is to be appreciated that various actions illustrated in FIG. 1 could occur substantially in parallel. By way of illustration, a first process could access a CT image of a region of tissue demonstrating cancerous pathology, a second process could extract texture features from the CT image, and a third process could classify the region as IA or AIS. While three processes are described, it is to be appreciated that a greater or lesser number of processes could be employed and that lightweight processes, regular processes, threads, and other approaches could be employed.

[0036] FIG. 2 illustrates a method 200 for combining a set of 3D histology reconstructions with a set of 3D CT images. Method 200 may be performed independently of method 100 or may be performed as an additional action during the performance of method 100. Method 200 includes, at 210, accessing a set of histology slices of a region of tissue demonstrating cancerous pathology. A member of the set of histology slices includes an invasive component and a non-invasive component. The invasive component may be invasive adenocarcinoma, while the non-invasive component may be in situ adenocarcinoma. The invasive component or the non-invasive component may be automatically annotated, or may be annotated by an expert pathologist.

[0037] Method 200 also includes, at 220, generating a 3D histology reconstruction of the set of histology slices. Generating the 3D histology reconstruction of the set of histology slices may include performing a group-wise registration of at least two members of the set of histology slices. Method 200 may determine a rotation or a translation with which to rotate or translate a first histology slice with respect to a second, different histology slice.

[0038] Method 200 also includes, at 230, accessing a set of 3D radiological images of the region of tissue. A member of the set of 3D radiological images includes a texture feature or a shape feature. In one embodiment, example methods and apparatus may extract at least 189 texture features from the set of 3D radiological images. In one embodiment, a member of the set of 3D radiological images is a CT image. In another embodiment, a member of the set of 3D radiological images may be an MR image.

[0039] In one embodiment, the 3D histology reconstruction may be refined using a local search constrained by the set of 3D radiological images. Refining the 3D histology reconstruction constrained by the set of 3D radiological images involves considering the GGN segmented by the CT imagery while adjusting the rotation or translation of the histology slices within the 3D histology reconstruction.

[0040] Method 200 also includes, at 240, identifying an optimal translation of the set of 3D radiological images relative to the 3D histology reconstruction. An optimal translation is a translation that matches or registers the 3D histology reconstruction to the 3D radiological images to within a threshold accuracy. Example methods and apparatus register the 3D histology reconstruction to the 3D radiological images to within at least 66% accuracy. In one embodiment, the optimal translation is an affine translation, or a deformable translation. Method 200 also includes, at 250, identifying an optimal rotation of the set of 3D radiological images relative to the 3D histology reconstruction.

[0041] Method 200 also includes, at 260, registering the 3D histology reconstruction with the set of 3D radiological images. The 3D histology reconstruction and the set of 3D radiological images may be registered based, at least in part, on the optimal translation or the optimal rotation.

[0042] Method 200 also includes, at 270, mapping the extent of cancerous invasion from the 3D histology reconstruction onto the set of 3D radiological images. Mapping the extent of cancerous invasion from the 3D histology reconstruction onto the set of 3D radiological images includes using an inverse of the optimized affine translation or an inverse of the optimized deformable translation.

[0043] Method 200 also includes, at 280, identifying, based, at least in part on the mapping, a texture feature or a shape feature associated with an invasive component. The texture feature associated with the invasive component may be a first-order statistic. The first-order statistic may be the top-ranked feature according to the Bhattacharyya distance. In one embodiment, the first-order statistic may be computed from a mean of CT intensity of a sliding 3×3×3 voxel window obtained from a member of the set of 3D CT images.

[0044] FIG. 3 illustrates a method 300 for distinguishing invasive cancerous pathology from non-invasive pathology. Method 300 includes, at 310, accessing a set of histology slices of a region of tissue demonstrating invasive cancerous pathology and non-invasive cancerous pathology. In one embodiment, the slices are stained with hematoxylin and eosin (H&E) and scanned at a 20× magnification with a pixel size of 0.25 μm. In another embodiment, other types of stain, scanning magnification, or pixel sizes may be employed.

[0045] Method 300 also includes, at 315, generating a 3D histology reconstruction of the set of histology slices. Generating the 3D histology reconstruction may include performing a group-wise registration of at least two members of the set of histology slices.

[0046] Method 300 also includes, at 320, accessing a set of 3D radiological images of the region of tissue. A member of the set of 3D radiological images includes a texture feature or a shape feature. In one embodiment, a member of the set of 3D radiological images is a CT image. In another embodiment, a member of the set of 3D radiological images is a DCE MRI image.

[0047] Method 300 also includes, at 325, registering the 3D histology reconstruction with the set of 3D radiological images. The 3D histology reconstruction and the set of 3D radiological images may be registered based, at least in part, on an optimal translation or an optimal rotation described in connection with actions 240-260 (FIG. 2). Registering the 3D histology reconstruction with the set of 3D radiological images may include performing an affine registration or an elastic registration.

[0048] Method 300 also includes, at 327, generating a mapping of invasive cancerous pathology from the 3D histology reconstruction onto the set of 3D radiological images based. In one embodiment, the mapping is based, at least in part, on the registration.

[0049] Method 300 also includes, at 330, identifying, in a member of the set of 3D radiological images, a texture feature or a shape feature associated with invasive cancerous pathology. The identification of the texture feature or shape feature may be based, at least in part, on the mapping.

[0050] Method 300 also includes, at 335, training a classifier to distinguish invasive cancerous pathology from non-invasive cancerous pathology. In one embodiment, the classifier may be trained to distinguish IA from AIS, based, at least in part, on the mapping, a texture feature, or a shape feature. Training a classifier may include training a support vector machine (SVM) using the “fused” histology-CT data as a set of training examples. The SVM may be trained to classify a region of tissue based on texture features associated with IA and to classify the region of tissue based on texture features associated with AIS. Example methods and apparatus may extract features from the “fused” histology-CT data and apply SVM regression to characterize the level of invasion. In one embodiment, SVM regression may be used to determine an IA score. For example, the SVM may learn to classify an image based on a texture feature that was extracted from the 3D radiological image that matches a previously learned texture feature to within a threshold. Other machine learning approaches may also be employed, including clustering, decision trees, or artificial neural networks.

[0051] In one embodiment, method 300 may include additional actions 340-380. In this embodiment method 300 includes, at 340 accessing a diagnostic 3D radiological image of a region of tissue demonstrating cancerous pathology. The diagnostic 3D radiological image may be a CT image or a DCE MRI image.

[0052] Method 300 also includes, at 350, extracting a texture feature or a shape feature from the diagnostic 3D radiological image. Extracting a texture feature or shape feature may be performed similarly to the extraction described in connection with action 120 (FIG. 1). The texture feature may be an intensity statistic, a Haralick feature, a Gabor feature, or other textural feature.

[0053] Method 300 also includes, at 360, providing the extracted texture feature or the extracted shape feature to the classifier. Providing the extracted texture feature of the extracted shape feature may be performed similarly to action 130 (FIG. 1). The extracted texture feature may be provided to a classifier across a network, to a classifier in a cloud service, or to a logic in an IA distinguishing apparatus.

[0054] Method 300 also includes, at 370, receiving, from the classifier, a first classification. The first classification may be based, at least in part, on the extracted texture feature or the extracted shape feature.

[0055] Method 300 also includes, at 380, distinguishing invasive cancerous pathology represented in the diagnostic 3D radiological image from non-invasive cancerous pathology represented in the diagnostic 3D image. Method 300 may distinguish IA from AIS represented in the diagnostic 3D radiological image based, at least in part, on the first classification and the texture feature or the shape feature.

[0056] In one example, a method may be implemented as computer executable instructions. Thus, in one example, a computer-readable storage device may store computer executable instructions that if executed by a machine (e.g., computer) cause the machine to perform methods described or claimed herein including method 100, method 200, and method 300. While executable instructions associated with the listed methods are described as being stored on a computer-readable storage device, it is to be appreciated that executable instructions associated with other example methods described or claimed herein may also be stored on a computer-readable storage device. In different embodiments, the example methods described herein may be triggered in different ways. In one embodiment, a method may be triggered manually by a user. In another example, a method may be triggered automatically.

[0057] FIG. 4 illustrates an example apparatus 400. Apparatus may be used for distinguishing invasive tissue from non-invasive tissue in a region of tissue demonstrating cancerous pathology. In one embodiment, apparatus 400 distinguishes IA from AIS in a region of tissue demonstrating cancerous pathology. Apparatus 400 includes a processor 410, a memory 420, a data store 430, an input/output (I/O) interface 440, a set of logics 460, and an interface 450 that connects the processor 410, the memory 420, the data store 430, the I/O interface 440, and the set of logics 460. Data store 430 stores a set of 3D radiological images of tissue demonstrating invasive or non-invasive pathology. In one embodiment, data store 430 stores a set of 3D radiological images of tissue demonstrating IA or AIS. A member of the set of 3D radiological images includes a texture feature or a shape feature.

[0058] The set of logics 460 includes a classification logic 462, a training logic 464, an image acquisition logic 466, and an IA prediction logic 468. In one embodiment, the functionality associated with the set of logics 460 may be performed, at least in part, by hardware logic components including, but not limited to, field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system on a chip systems (SOCs), or complex programmable logic devices (CPLDs). In one embodiment, individual members of the set of logics 460 are implemented as ASICs or SOCs.

[0059] In one embodiment, classification logic 462 generates the invasiveness classification by distinguishing IA from AIS represented in the diagnostic 3D radiological image based, at least in part, on a texture feature extracted from the diagnostic 3D radiological image or a shape feature extracted from the diagnostic 3D radiological image. In another embodiment, classification logic 462 may distinguish invasion from non-invasion associated with other types of cancer, including breast cancer.

[0060] In one embodiment, training logic 464 trains the classification logic 462 using a 3D histology reconstruction of a region of tissue demonstrating IA or AIS combined with the set of 3D radiological images of the region of tissue. Training logic 464 combines the 3D histology reconstruction with the set of 3D radiological images by accessing a set of histology slices of a region of tissue demonstrating IA or AIS. Training logic 464 may then generate a 3D histology reconstruction of the set of histology slices. Training logic 464 may then access the set of 3D radiological images of the region of tissue. Training logic 464 may then register the 3D histology reconstruction with the set of 3D radiological images. Training logic 464 may generate a mapping of IA from the 3D histology reconstruction onto the set of 3D radiological images, and identify, in a member of the set of 3D radiological images, a texture feature or a shape feature associated with IA based, at least in part, on the mapping.

[0061] In one embodiment, the set of logics 460 also includes a display logic. The display logic may control a computer assisted diagnostic (CADx) system to display the invasiveness classification, the diagnostic 3D radiological image, the texture feature, the shape feature, or the 3D histology reconstruction combined with the set of 3D radiological images. By displaying the diagnostic 3D radiological image along with the invasiveness classification, the texture feature, the shape feature, or the 3D histology reconstruction combined with the set of 3D radiological images, example apparatus provide a timely and intuitive way for a human pathologist to more accurately classify pathologies demonstrated by a patient, thus improving on conventional approaches to characterizing cancerous invasion, or predicting cancer recurrence and disease progression.

[0062] FIG. 5 illustrates an example computer 500 in which example methods illustrated herein can operate and in which example logics may be implemented. In different examples, computer 500 may be part of a CT system or an MRI system, may be operably connectable to a CT system or an MRI system, or may be part of a CADx system.

[0063] Computer 500 includes a processor 502, a memory 504, and input/output ports 510 operably connected by a bus 508. In one example, computer 500 may include a set of logics 530 that perform a method of distinguishing IA from AIS in a region of tissue demonstrating cancerous pathology. Thus, the set of logics 530, whether implemented in computer 500 as hardware, firmware, and/or a combination thereof may provide means (e.g., circuits, hardware) for distinguishing IA from AIS in a region of tissue demonstrating cancerous pathology. In different examples, the set of logics 530 may be permanently and/or removably attached to computer 500. In one embodiment, the functionality associated with the set of logics 530 may be performed, at least in part, by hardware logic components including, but not limited to, field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system on a chip systems (SOCs), or complex programmable logic devices (CPLDs). In one embodiment, individual members of the set of logics 530 are implemented as ASICs or SOCs.

[0064] Processor 502 can be a variety of various processors including dual microprocessor and other multi-processor architectures. Memory 504 can include volatile memory and/or non-volatile memory. A disk 506 may be operably connected to computer 500 via, for example, an input/output interface (e.g., card, device) 518 and an input/output port 510. Disk 506 may include, but is not limited to, devices like a magnetic disk drive, a tape drive, a Zip drive, a flash memory card, or a memory stick. Furthermore, disk 506 may include optical drives like a CD-ROM or a digital video ROM drive (DVD ROM). Memory 504 can store processes 514 or data 517, for example. Disk 506 or memory 504 can store an operating system that controls and allocates resources of computer 500.

[0065] Bus 508 can be a single internal bus interconnect architecture or other bus or mesh architectures. While a single bus is illustrated, it is to be appreciated that computer 500 may communicate with various devices, logics, and peripherals using other busses that are not illustrated (e.g., PCIE, SATA, Infiniband, 1394, USB, Ethernet).

[0066] Computer 500 may interact with input/output devices via I/O interfaces 518 and input/output ports 510. Input/output devices can include, but are not limited to, digital whole slide scanners, an optical microscope, a keyboard, a microphone, a pointing and selection device, cameras, video cards, displays, disk 506, network devices 520, or other devices. Input/output ports 510 can include but are not limited to, serial ports, parallel ports, or USB ports.

[0067] Computer 500 may operate in a network environment and thus may be connected to network devices 520 via I/O interfaces 518 or I/O ports 510. Through the network devices 520, computer 500 may interact with a network. Through the network, computer 500 may be logically connected to remote computers. The networks with which computer 500 may interact include, but are not limited to, a local area network (LAN), a wide area network (WAN), or other networks.

[0068] References to “one embodiment”, “an embodiment”, “one example”, and “an example” indicate that the embodiment(s) or example(s) so described may include a particular feature, structure, characteristic, property, element, or limitation, but that not every embodiment or example necessarily includes that particular feature, structure, characteristic, property, element or limitation. Furthermore, repeated use of the phrase “in one embodiment” does not necessarily refer to the same embodiment, though it may.

[0069] “Computer-readable storage device”, as used herein, refers to a device that stores instructions or data. “Computer-readable storage device” does not refer to propagated signals. A computer-readable storage device may take forms, including, but not limited to, non-volatile media, and volatile media. Non-volatile media may include, for example, optical disks, magnetic disks, tapes, and other media. Volatile media may include, for example, semiconductor memories, dynamic memory, and other media. Common forms of a computer-readable storage device may include, but are not limited to, a floppy disk, a flexible disk, a hard disk, a magnetic tape, other magnetic device, an application specific integrated circuit (ASIC), a compact disk (CD), other optical device, a random access memory (RAM), a read only memory (ROM), a memory chip or card, a memory stick, and other media or device from which a computer, a processor or other electronic device can read.

[0070] “Logic”, as used herein, includes but is not limited to hardware, or firmware, or combinations of each to perform a function(s) or an action(s), or to cause a function or action from another logic, method, or system. Logic may include a software controlled microprocessor, a discrete logic (e.g., ASIC), an analog circuit, a digital circuit, a programmed logic device, a memory device containing instructions, and other physical devices. Logic may include one or more gates, combinations of gates, or other circuit components. Where multiple logical logics are described, it may be possible to incorporate the multiple logical logics into one physical logic. Similarly, where a single logical logic is described, it may be possible to distribute that single logical logic between multiple physical logics.

[0071] To the extent that the term “includes” or “including” is employed in the detailed description or the claims, it is intended to be inclusive in a manner similar to the term “comprising” as that term is interpreted when employed as a transitional word in a claim.

[0072] Throughout this specification and the claims that follow, unless the context requires otherwise, the words ‘comprise’ and ‘include’ and variations such as ‘comprising’ and ‘including’ will be understood to be terms of inclusion and not exclusion. For example, when such terms are used to refer to a stated integer or group of integers, such terms do not imply the exclusion of any other integer or group of integers.

[0073] To the extent that the term “or” is employed in the detailed description or claims (e.g., A or B) it is intended to mean “A or B or both”. When the applicants intend to indicate “only A or B but not both” then the term “only A or B but not both” will be employed. Thus, use of the term “or” herein is the inclusive, and not the exclusive use. See, Bryan A. Garner, A Dictionary of Modern Legal Usage 624 (2d. Ed. 1995).

[0074] While example systems, methods, and other embodiments have been illustrated by describing examples, and while the examples have been described in considerable detail, it is not the intention of the applicants to restrict or in any way limit the scope of the appended claims to such detail. It is, of course, not possible to describe every conceivable combination of components or methodologies for purposes of describing the systems, methods, and other embodiments described herein. Therefore, the invention is not limited to the specific details, the representative apparatus, and illustrative examples shown and described. Thus, this application is intended to embrace alterations, modifications, and variations that fall within the scope of the appended claims.