Description:FIELD OF THE INVENTION
[0001] The invention relates in general to systems and methods for use in the identification of oral pre-cancerous and cancerous conditions. More particularly, the present invention relates to an oral cancer screening device for use in the identification of oral pre-cancerous and cancerous conditions.

BACKGROUND OF THE INVENTION
[0002] Cancer is one of the leading causes of deaths globally and oral cancer is the sixth most common cancers in Asia and ranks top three of all cancers in India. Oral cancer occurs more often in people from the lower end of the socioeconomic scale and India has one third of the world's oral cancer cases. When identified at an early stage by screening larger populations, their progression can be controlled which can reduce the incidence of oral cancer. Regular screening and early diagnosis of cancers greatly increases the chances for successful treatment and enhances the healthcare outcomes. A number of point-of-care and portable devices are being developed and used in the clinics for such screening purposes for determining a patient risk for oral cancer.

[0003] However, such methods are not reliable and need expensive machinery to perform the screening. Hence, there exists a need for a simple, inexpensive screening mechanism for determining a patient risk for oral cancer.

SUMMARY OF THE INVENTION
[0004] In one embodiment, the invention described herein can be used as an oral screening device for identifying indicators for oral cancer. The screening tool is configured for empowering a dentist to perform the screening in an effective manner for identifying precancerous and cancerous conditions. In another embodiment, a method for performing screening for identifying precancerous and cancerous conditions is described.

[0005] Accordingly, a system and method for detecting indicators for oral cancer in a subject are described. The system comprises at least one processing unit 102; and one or more memory unit 103s configured to store software instructions configured for execution by the at least one processing unit 102 in order to cause the system to: receive an image of the oral cavity of a patient; identify a plurality of indicators throughout the image of the oral cavity using a convolutional network; detect a presence or absence of a pre-cancerous or cancerous tissue based the identified indicators using the convolutional network; and generate an indicator representing the severity of the pre-cancerous or cancerous tissue based on the presence or absence of the pre-cancerous or cancerous tissue using the convolutional network.

[0006] In another embodiment, a method of screening oral cancer, the method comprising steps of capturing at least one image of the oral cavity of a subject, using an imaging device 101, processing the image for identifying probable presence of a pre-cancerous or a cancerous tissue, by a processing unit 102 and wherein the processing comprises pre-processing the image, identifying a plurality of indicators throughout the image using a convolutional network, generating an indicator depicting the probability of occurrence of oral cancer in the subject, based on the identified indicators using the convolutional network and displaying the indicator on a display unit 104 coupled to the processing unit 102.

[0007] In yet another embodiment, a computer implemented method of screening oral cancer, the method comprising steps of providing a memory unit 103 and a processing unit 102, the memory unit 103 configured for storing instructions to be executed by the processing unit 102, which when executed cause the processing unit 102 to capture at least one image of the oral cavity of a subject, using an imaging device 101, process the image for identifying probable presence of an oral cancer and wherein the processing comprises pre-process the image, identify a plurality of indicators throughout the image using a convolutional network, generate an indicator depicting the probability of occurrence of oral cancer in the subject, based on the identified indicators using the convolutional network and display the primary index on a display unit 104 coupled to the processing unit 102.

[0008] The foregoing has outlined rather broadly several embodiments of the present invention in order that the detailed description of the invention that follows may be better understood. Additional features and advantages of the invention will be described hereinafter which form the subject of the claims of the invention. It should be appreciated by those skilled in the art that the conception and the specific embodiment disclosed might be readily utilized as a basis for modifying or redesigning the structures for carrying out the same purposes as the invention. It should be realized by those skilled in the art that such equivalent constructions do not depart from the spirit and scope of the invention as set forth in the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS
[0009] For a more complete understanding of the present invention, and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:

[0010] FIG. 1 is a schematic view illustrating the basic components of the oral screening device and their interrelationship.

[0011] FIG. 2 is a flow diagram illustrating the method of performing oral screening to identify indicators for pre-cancerous and cancerous tissues.

DETAILED DESCRIPTION OF THE INVENTION

[0012] Figure 1 illustrates a block diagram of a system 1000 in accordance with the invention. The system 1000 comprises at least one processing unit 102; and one or more memory units 103 configured to store software instructions configured for execution by the at least one processing unit 102 in order to cause the system 1000 to: receive an image of the oral cavity of a patient; identify a plurality of indicators throughout the image of the oral cavity using a convolutional network; detect a presence or absence of a pre-cancerous or cancerous tissue based the identified indicators using the convolutional network; and generate an indicator representing the severity of the pre-cancerous or cancerous tissue based on the presence or absence of the pre-cancerous or cancerous tissue using the convolutional network.

[0013] The image of the oral cavity, herein, refers to a two-dimensional array of digital image data, however, this is merely illustrative and not limiting of the scope of the invention. The one or more memory units 103 is, for example, a database to store a structured collection of data. In an embodiment, the one or more memory units 103 may be an internal part of the system 1000. In another embodiment, the one or more memory units 103 may be remotely located and accessed via a network. The one or more memory units 103 may be, for example, removable and/or non-removable data storage such as a tape, a magnetic disk, an optical disks, a flash memory card, etc. The one or more memory units 103 may comprise, for example, random access memory (RAM), read only memory (ROM), electrically erasable programmable read-only memory (EEPROM), a digital versatile disks (DVD), a compact disk (CD), a flash memory, a magnetic tape, a magnetic disk storage, or any combination thereof that can be used to store and access information and is a part of the system 1000.

[0014] The abnormality is one of a lesion like an aphthous ulcer, reticular oral lichen planus, koplakia, erythroplakia, erythroplakia, oral squamous cell carcinoma, or the like. The severity of the pre-cancerous or cancerous tissue are represented as levels of increasing seriousness of the pre-cancerous or cancerous tissue, which is indicated by an indicator. The indicator represents a risk score allowing a user to distinguish at least the following: i) benign lesions, ii) dysplastic lesions, and iii) cancerous lesions.

[0015] The processing unit 102 of the system 1000 receives a reference dataset from one or more input devices. The reference dataset comprises a plurality of image of the oral cavities. Hereafter, the image of the oral cavities in the reference dataset are referred to as reference image of the oral cavities.

[0016] The processing unit 102 is configured for receiving the image of the oral cavity from an imaging device 101. For such uses the imaging device 101 is adapted as a handheld portable device that allows for accurate but easy oral scanning equipment that can be easily taken to remote geographies where medical facilities may be scarce. For the implementations with a camera module, digital images may be taken by relatively low skilled personnel and the images be stored and may be sent through an appropriate communication means like email, internet, mobile communication to a remote physician, clinic or a hospital for further reading, analysis and reporting.

[0017] Accordingly, in one embodiment, a personal digital assistant is configured to obtain images of the oral cavity. The images are processed to identify the indicators for pre-cancerous and cancerous tissues. For this the computer implemented method can be downloaded on the processing unit 102 of the imaging device 101. The imaging device 101 for example is a personal digital assistant such as a desktop, a laptop, a smart phone and a tablet. Further, the system 1000 stores the reference dataset in the one or more memory units 103 of the system 1000.

[0018] The reference image of the oral cavity is a two-dimensional array of digital image data used for the purpose of training the system 1000. In this invention, the term ‘training’ generally refers to a process of developing the system 1000 for the detection and classification of the pre-cancerous or cancerous tissue based the reference dataset and a reference ground- truth file. The reference ground-truth file comprises a label and a reference image of the oral cavity identifier for each of the reference image of the oral cavity. The label provides information about the reference image of the oral cavity such as a presence or absence of a pre-cancerous or cancerous tissue, the type of pre-cancerous or cancerous tissue and the corresponding severity of the pre-cancerous or cancerous tissue identified in the reference image of the oral cavity. The reference image of the oral cavity identifier of a reference image of the oral cavity is, for example, a name or identity assigned to the reference image of the oral cavity.

[0019] In an embodiment, an annotator annotates each of the reference images using an annotation platform of the system 1000. The annotation platform is a graphical user interface (GUI) provided for the annotator to interact with the system 1000. The annotator accesses the reference image of the oral cavities via the annotation platform. The annotator creates a label with information about a presence or absence of a pre-cancerous or cancerous tissue, the type of pre-cancerous or cancerous tissue and the corresponding severity of the pre-cancerous or cancerous tissue based on the annotation. The annotator is usually a trained/certified specialist in accurately annotating the image of the oral cavity by analyzing the indicators present in the reference image of the oral cavity. In an example, consider that the annotator annotates the reference image of the oral cavities for the pre-cancerous or cancerous tissue - The label of the reference image of the oral cavity represents the OC (Oral Cavity) severity level associated with the patient.

[0020] For example, the annotator labels each of the reference image of the oral cavity as one of five severity classes- ‘No OC’, ‘OC1’, ‘OC2’, ‘OC3’ and ‘OC4’ based on an increasing seriousness of OC. Here, ‘No OC’, ‘OC1’, ‘OC2’, ‘OC3’ and ‘OC4’ represents the labels indicating different levels of increasing severity of OC associated with the patient. The annotator analyses the indicators in the image of the oral cavity and accordingly marks the label. If the annotator detects a microaneurysm, then the annotator considers it as a mild level of OC and marks the label as OC1 for the reference image of the oral cavity. Similarly, if the annotator detects one or more of the following – a hard exudate, a soft exudate, a hemorrhage, a venous loop, a venous beading, etc., then the annotator marks the label as OC2 for the reference image of the oral cavity. The label OC2 indicates a moderate level of OC. The annotator marks the label as OC3 for the reference image of the oral cavity with a severe level of OC upon detection of multiple hemorrhages, hard or soft exudates, etc., and OC4 for the reference image of the oral cavity with a proliferative level of OC upon detection of vitreous hemorrhage, neovascularization, etc. The reference image of the oral cavity with no traces of OC is marked with the label as ‘No OC’ by the annotator.

[0021] The annotator stores the label and the reference image of the oral cavity identifier for each of reference image of the oral cavity in the reference ground-truth file located in the one or more memory units 103. The label provides information about the type of pre-cancerous or cancerous tissue and the corresponding severity of the pre-cancerous or cancerous tissue as annotated by the annotator. The reference image of the oral cavity identifier of a reference image of the oral cavity is, for example, a name or identity assigned to the reference image of the oral cavity.

[0022] In another embodiment, the processing unit 102 identifies the indicators throughout each of the reference image of the oral cavity to detect the presence or absence of the pre-cancerous or cancerous tissue using image analysis techniques. The processing unit 102 classifies the severity of the pre-cancerous or cancerous tissue based on the presence of the pre-cancerous or cancerous tissue using a set of predetermined rules. The predetermined rules comprise considering a type of each of the indicators, a count of each indicators, a region of occurrence of each of the indicators, a contrast level of each of the indicators, a size of each of the indicators or any combination thereof to recognize the pre-cancerous or cancerous tissue and the severity of the pre-cancerous or cancerous tissue. The processing unit 102 classifies each of the detected pre-cancerous or cancerous tissues according to a corresponding severity grading and generates the label. The processing unit 102 communicates with the one or more memory units 103 to store the label and the reference image of the oral cavity identifier for each of reference image of the oral cavity in the reference ground-truth file.

[0023] The processing unit 102 utilizes the reference dataset to train the convolutional network for subsequent detection and classification of the oral cancer disease in the image of the oral cavity. Hereafter, the image of the oral cavity which is subsequently analyzed by the processing unit 102 is referred to as an input image of the oral cavity for clarity.

[0024] The processing unit 102 pre-processes each of the reference image of the oral cavities. For each of the reference image of the oral cavity, the processing unit 102 executes the following steps as part of the pre-processing. The processing unit 102 separates any text matter present at the border of the reference image of the oral cavity. The processing unit 102 adds a border to the reference image of the oral cavity with border pixel values as zero. The processing unit 102 increases the size of the reference image of the oral cavity by a predefined number of pixels, for example, 20 pixels width and height. The additional pixels added are of a zero value. The processing unit 102 next converts the reference image of the oral cavity from a RGB color image to a grayscale image. The processing unit 102 now binarize the reference image of the oral cavity using histogram analysis. The processing unit 102 applies repetitive morphological dilation with a rectangular element of size [5, 5] to smoothen the binarized reference image of the oral cavity. The processing unit 102 acquires all connected regions such as, text matter of the smoothen reference image of the oral cavity to separate text matter present in the reference image of the oral cavity from a foreground image.

[0025] Once the processing unit 102 identifies the abnormality in the reference image of the oral cavity, the processing unit 102 further blurs the reference image of the oral cavity using a Gaussian filter. The processing unit 102 compares an image width and an image height of the blurred reference image of the oral cavity based on Equation 1.

[0026] Image width > 1.2(image height) -------Equation 1

[0027] The processing unit 102 calculates a maximum pixel value of a left half, a maximum pixel value of a right half and a maximum background pixel value for the blurred reference image of the oral cavity when the image width and the image height of the blurred identified satisfies the Equation 1. The maximum background pixel value (Max_background pixel value) is given by the below Equation 2. The term ‘max_pixel_left’ in Equation 2 is the maximum pixel value of the left half of the blurred identified abnormality. The term ‘max_pixel_right’ in Equation 2 is the maximum pixel value of the right half of the blurred reference image of the oral cavity.

[0028] Max_background pixel value = max(max_pixel_left, max_pixel_right) ------Equation 2

[0029] The processing unit 102 further extracts foreground pixel values from the blurred reference image of the oral cavity by considering pixel values which satisfy the below Equation 3.

[0030] All pixel values > max_background_pixel_value + 10 Equation 3

[0031] The processing unit 102 calculates a bounding box using the extracted foreground pixel values from the blurred reference image of the oral cavity. The processing unit 102 processes the bounding box to obtain a resized image using cubic interpolation of shape, for example, [256, 256, 3]. The reference image of the oral cavity at this stage is referred to as the pre-processed reference image of the oral cavity. The processing unit 102 stores the pre-processed reference image of the oral cavities in a pre-processed reference dataset. The ground-truth file associated with the reference dataset holds good even from the pre-processed reference dataset. The processing unit 102 stores the pre-processed reference dataset in the one or more memory units 103.

[0032] The processing unit 102 splits the pre-processed reference dataset into two sets – a training set and a validation set. Hereafter, the pre-processed reference image of the oral cavities in the training set is termed as training image of the oral cavities and the pre-processed reference image of the oral cavities in the validation set is termed as validation image of the oral cavities for simplicity. The training set is used to train the convolutional network to assess the training image of the oral cavities based on the label associated with each of the training image of the oral cavity. The validation set is typically used to test the accuracy of the convolutional network.

[0033] The processing unit 102 augments the training images in the training set. The processing unit 102 preforms the following steps for the augmentation of the training set. The processing unit 102 randomly shuffles the training image of the oral cavities to divide the training set into a plurality of batches. Each batch is a collection of a predefined number of training image of the oral cavities. The processing unit 102 randomly samples each batch of training image of the oral cavities. The processing unit 102 processes each batch of the training image of the oral cavities using affine transformations. The processing unit 102 translates and rotates the training images in the batch randomly based on a coin flip analogy. The processing unit 102 also adjusts the color and brightness of each of the training image of the oral cavities in the batch randomly based on the results of the coin flip analogy.

[0034] The processing unit 102 trains the system 1000 using the batches of augmented training image of the oral cavities via the convolutional network. In general, the convolutional network is a class of deep artificial neural networks that can be applied to analyzing visual imagery. The general arrangement of the convolutional network is as follows. The convolutional network comprising ‘n’ convolutional stacks applies a convolution operation to the input and passes an intermediate result to a next layer. Each convolutional stack comprises a plurality of convolutional layers. A first convolution stack is configured to convolve pixels from an input with a plurality of filters to generate a first indicator map. The first convolutional stack also comprises a first subsampling layer configured to reduce a size and variation of the first indicator map. The first convolutional layer of the first convolutional stack is configured to convolve pixels from the input with a plurality of filters. The first convolutional stack passes an intermediate result to the next layer. Similarly, each convolutional stack comprises a sub-sampling layer configured to reduce a size (width and height) of the indicators stack. The input is analyzed based on reference data to provide a corresponding output.

[0035] Similar to the training set, the processing unit 102 groups the validation images of the validation set into a plurality of batches. Each batch comprises multiple validation image of the oral cavities. The processing unit 102 validates each of the validation image of the oral cavities in each batch of the validation set using the convolutional network. The processing unit 102 compares a result of the validation against a corresponding label of the validation image of the oral cavity by referring to the reference ground-truth file. The processing unit 102 thus evaluates a convolutional network performance of the convolutional network for the batch of validation set.

[0036] The processing unit 102 optimizes the convolutional network parameters using an optimizer, for example, a Nadam optimizer which is an Adam optimizer with Nesterov Momentum. The optimizer iteratively optimizes the parameters of the convolutional network during multiple iterations using the training set. Here, each iteration refers to a batch of the training set. The processing unit 102 evaluates a convolutional network performance of the convolutional network after a predefined number of iterations on the validation set. Here, each iteration refers to a batch of the validation set.

[0037] Thus, the processing unit 102 trains the convolutional network based on the augmented training set and tests the convolutional network based on the validation set. Upon completion of training and validation of the convolution network based on the convolutional network performance, the system 1000 is ready to assess the input image of the oral cavity based on the indicators present in the input image of the oral cavity.

[0038] The processing unit 102 of the system 1000 receives the input image of the oral cavity from the imaging device. The processing unit 102 pre-processes the input image of the oral cavity similar to that of the reference image. The processing unit 102 test-time augments the preprocessed input image of the oral cavity. The processing unit 102 performs the following steps to test- time augment the preprocessed input image of the oral cavity. The processing unit 102 converts the preprocessed input image of the oral cavity into a plurality of test time images, for example, twenty test time images, using deterministic augmentation. The processing unit 102 follows the same steps to augment the input image of the oral cavity as that of the training image of the oral cavity, except that the augmentations are deterministic. Thus, the processing unit 102 generates deterministically augmented twenty test time images of the preprocessed input image of the oral cavity. The processing unit 102 processes the deterministically augmented twenty test time images of the preprocessed input image of the oral cavity using the convolutional network comprising ‘n’ convolutional stacks. The predicted probabilities of the twenty test time images are averaged over to get a final prediction result. The final prediction result provides a probability value for each of the pre-cancerous or cancerous tissue and a corresponding pre-cancerous or cancerous tissue severity level associated with the input image of the oral cavity. The probability value is an indication of a confidence that identified indicators are of a particular pre-cancerous or cancerous tissue and a corresponding severity of the pre-cancerous or cancerous tissue. The output indicates a presence or absence of a pre-cancerous or cancerous tissue and related severity of the pre-cancerous or cancerous tissue associated with the input image of the oral cavity.

[0039] In one embodiment, the deterministically augmented twenty test time images of the preprocessed input image of the oral cavity is the input to a first convolutional stack (CS1) of the convolutional network. Each of the deterministically augmented twenty test time images are processed by the convolutional network.

[0040] The deterministically augmented test time image is, for example, represented as a matrix of width 448 pixels and height 448 pixels with ‘3’ channels. That is, the deterministically augmented test time image is a representative array of pixel values is 448 x 448 x 3. The input to the first convolutional stack (CS1) is a color image of size 448 x 448. The first convolution stack (CS1) comprises the following sublayers - a first convolutional layer, a first subsampling layer, a second convolutional layer, a third convolutional layer and a second subsampling layer in the same order. The output of a sublayer is an input to a consecutive sublayer. In general, a subsampling layer is configured to reduce a size and variation of its input and a convolutional layer convolves its input with a plurality of filters, for example, filters of size 3x3. The output of the first convolutional stack (CS1) is a reduced image represented as a matrix of width 112 pixels and height 112 pixels with n1 channels. That is, the output of the first convolutional stack (CS1) is a representative array of pixel values 112 x 112 x n1.

[0041] This is the input to a second convolutional stack (CS2). The second convolutional stack (CS2) comprises the following sublayers - four convolutional layers and a subsampling layer arranged in the same order. Again, the output of a sublayer is an input to a consecutive sublayer. The second convolutional stack (CS2) convolves the representative array of pixel values 112 x 112 x n1 and reduces it to a representative array of pixel values of 56 x 56 x n2. The representative array of pixel values of 56 x 56 x n2 is an input to a third convolutional stack (CS3).

[0042] The third convolutional stack (CS3) comprises the following sublayers - four convolutional layers and a subsampling layer arranged in the same order. Again, the output of a sublayer is an input to a consecutive sublayer. The third convolutional stack (CS3) convolves the representative array of pixel values 56 x 56 x n2 and reduces it to a representative array of pixel values of 28 x 28 x n3. The representative array of pixel values of 28 x 28 x n3 is an input to a fourth convolutional stack (CS4).

[0043] The fourth convolutional stack (CS4) comprises the following sublayers -four convolutional layers and a subsampling layer arranged in the same order. Again, the output of a sublayer is an input to a consecutive sublayer. The fourth convolutional stack (CS4) convolves the representative array of pixel values 28 x 28 x n3 and reduces it to a representative array of pixel values of 14 x 14 x n4. The representative array of pixel values of 14 x 14 x n4 is an input to a fifth convolutional stack (CS4).

[0044] The fifth convolutional stack (CS5) comprises the following sublayers - four convolutional layers and a subsampling layer arranged in the same order. Again, the output of a sublayer is an input to a consecutive sublayer. The fifth convolutional stack (CS5) convolves the representative array of pixel values 14 x 14 x n4 and reduces it to a representative array of pixel values of 7 x 7 x n5. The representative array of pixel values of 7 x 7 x n5 is a first input to a concatenation block (C).

[0045] The output of the third convolutional stack (CS3) is an input to a first subsampling block (SS1). The representative array of pixel values of 28 x 28 x n3 is the input to the first subsampling block (SS1). The first subsampling block (SS1) reduces the input with a stride of 4 to obtain an output of a representative array of pixel with value of 7 x 7 x n3. This is a second input to the concatenation block (C).

[0046] The output of the fourth convolutional stack (CS4) is an input to a second subsampling block (SS2). The representative array of pixel values of 14 x 14 x n4 is the input to the second subsampling block (SS2). The second subsampling block (SS2) reduces the input with a stride of 2 to obtain an output of a representative array of pixel with value of 7 x 7 x n4. This is a third input to the concatenation block (C).

[0047] The concatenation block (C) receives the first input from the fifth convolutional stack (CS5), the second input from the first subsampling block (SS1) and the third input from the second subsampling block (SS2). The concatenation block (C) concatenates the three inputs received to generate an output of value 7 x 7 x (n5 + n4 + n3). The output of the concatenation block (C) is an input to a probability block (P).
[0048] The probability block (P) provides a probability of the presence or absence of the pre-cancerous or cancerous tissue and related severity of the pre-cancerous or cancerous tissue. The predicted probabilities of the twenty test time images are averaged to get a final prediction result. The output of the convolutional network provides a probability value for each of the pre-cancerous or cancerous tissue and a corresponding pre-cancerous or cancerous tissue severity level associated with the input image of the oral cavity. The probability block (P) as shown in the Figure 2 provides five values by considering the pre-cancerous or cancerous tissue to be OC. The output of the probability block are five values depicting the probability for each OC severity level – DR0 (no OC), OC1 (mild OC level), OC2 (moderate OC level), OC3 (severe OC level) and OC4 (proliferative OC level).

[0049] The system 1000 considers an image capture device characteristics of an image capture device also as one of the parameters to assess the input image of the oral cavity. The image capture device characteristics is a resolution, an illumination factor, a field of view or any combination thereof. The image capture device is, for example, a camera, a camera attached to a smartphone, etc., used to capture the input image of the oral cavity. For example, the image capture device is the input device used to capture the input image of the oral cavity.

[0050] In an embodiment, the processing unit 102 of the system 1000 considers a manufacturer and version of the image capture device to determine a predefined score for the image capture device characteristics of the image capture device. This predefined score for the image capture device characteristics is used to assess the input image of the oral cavity. The predefined score for the image capture device characteristics denotes a superiority of the image capture device characteristics. The predefined score for the image capture device characteristics is a numeric value within the range of [0, 1]. Here, 0 defines a least value and 1 defines a highest value of the predefined score for the image capture device characteristics. For example, the predefined score for the image capture device characteristics for multiple manufacturers of image capture device is initially stored in the one or more memory units 103 by an operator of the system 1000. By considering the image capture device characteristics of an image capture device to assess the quality of the input image of the oral cavity, the flexibility of the system 1000 is increased, thereby providing customized results for the input image of the oral cavity captured using the image capture device of multiple manufacturers. Thus, the processing unit 102 considers the output of the convolutional network and the image capture device characteristics of the image capture device to determine the presence or absence of a pre-cancerous or cancerous tissue and/or related severity of the pre-cancerous or cancerous tissue associated with the input image of the oral cavity. The processing unit 102 thereby considers the quality of the input image of the oral cavity along with the output of the convolutional network to determine the presence or absence of a pre-cancerous or cancerous tissue and/or related severity of the pre-cancerous or cancerous tissue associated with the input image of the oral cavity.

[0051] The processing unit 102 displays the presence or absence of a pre-cancerous or cancerous tissue and/or related severity of the pre-cancerous or cancerous tissue associated with the input image of the oral cavity via a display 104 to a user. For example, suitable suggestions with a set of instructions to the user may also be included and provided via a pop-up box displayed on a screen. The system 1000 may also generate a report comprising the input image of the oral cavity, the type of the pre-cancerous or cancerous tissue and the severity of the pre-cancerous or cancerous tissue and communicated to the patient via an electronic mail. The report could also be stored in the one or more memory units 103 of the system 1000.

[0052] In an embodiment, the processing unit 102 assesses a quality measure of each of the reference image of the oral cavities in the reference dataset. The quality measure is also stored as a part of the label associated with the reference image of the oral cavity. The processing unit 102 trains the convolutional network to learn the quality measure of the reference image of the oral cavity along with the identification of the indicators in the reference image of the oral cavity. The processing unit 102 assesses the input image of the oral cavity based on the training. The processing unit 102 identifies the quality measure of the input image of the oral cavity using the convolutional network. The processing unit 102 may also refer to a user defined threshold to define the quality measure of the input image of the oral cavity. The user defined threshold is user defined to increase a flexibility of the system 1000. The user defined threshold is the variable factor which may be used to vary the quality measure of the input image of the oral cavity to conveniently suit the requirements of the user, for example, medical practitioner. The user defined threshold may be varied to vary the quality measure of the input image of the oral cavity based on the doctor’s grading experience. The system 1000 may further display 104 a message to an operator to retake another image of the oral cavity of the patient when the quality measure of the input image of the oral cavity is below a threshold. The system 1000 may further consider characteristics of a device used to capture the input image of the oral cavity of the patient as an additional parameter to assess the quality measure of the input image of the oral cavity.

[0053] Figure 2 illustrates a flowchart for determination of the presence or absence of the pre-cancerous or cancerous tissue and related severity of the pre-cancerous or cancerous tissue associated with the input image of the oral cavity in accordance with the invention. At step 201, the processing unit 102 receives the image of the oral cavity of the patient. The image of the oral cavity is the input image of the oral cavity. The input image of the oral cavity is a two-dimensional array of digital image data. At step 302, the processing unit 102 identifies multiple indicators throughout the image of the oral cavity using the convolutional network.

[0054] At step 202, the processing unit 102 is configured for processing the image for identifying probable presence of a pre-cancerous or a cancerous tissue, detects the presence or absence of the pre-cancerous or cancerous tissue based on the identified indicators using the convolutional network. The step 202 includes pre-processing the image at step 203, identifying a plurality of indicators throughout the image using a convolutional network at step 204 and generating an indicator depicting the probability of occurrence of oral cancer in the subject at step 205, based on the identified indicators using the convolutional network.

[0055] The indicator represents the severity level of the pre-cancerous or cancerous tissue. At step 304, the processing unit 102 classifies the severity of the pre-cancerous or cancerous tissue based on the presence or absence of the pre-cancerous or cancerous tissue using the convolutional network. The severity of the pre-cancerous or cancerous tissue may be classified into several levels depending upon the severity. The general concepts of the current invention are not limited to a particular number of severity levels. In an embodiment, one severity level could be used which satisfies only the detection of the oral cancer disease. In another embodiment, multiple severity levels could be used to classify the pre-cancerous or cancerous tissue. In another embodiment, multiple pre-cancerous or cancerous tissues could be detected based on the identified indicators. The system 1000 classifies each of the detected pre-cancerous or cancerous tissues based on the severity.
[0056] The method further includes displaying the indicator on a display unit 104 coupled to the processing unit 102 at step 210.

[0057] In yet another embodiment, a computer implemented method of screening oral cancer, the method comprising steps of providing a memory unit 103 and a processing unit 102, the memory unit 103 configured for storing instructions to be executed by the processing unit 102, which when executed cause the processing unit 102 to capture at least one image of the oral cavity of a subject, using an imaging device 101, process the image for identifying probable presence of an oral cancer and wherein the processing comprises pre-process the image, identify a plurality of indicators throughout the image using a convolutional network, generate an indicator depicting the probability of occurrence of oral cancer in the subject, based on the identified indicators using the convolutional network and display the primary index on a display unit 104 coupled to the processing unit 102.

[0058] The system 1000 using the convolutional network emphases on classifying the entire image of the oral cavity as a whole. This improves efficiency and reduces errors in identifying various medical conditions. The system 1000 acts as an important tool in the detection, monitoring a progression of a pre-cancerous or cancerous tissue and/or or a response to a therapy. The system 1000 trains the convolutional network to detect all indicative indicators related to multiple pre-cancerous or cancerous tissues. The system 1000 accurately detects indicators throughout the input image of the oral cavity which are indicative of disease conditions to properly distinguish indicators of a healthy oral cavity from indicators which define pre-cancerous or cancerous tissues.

[0059] The system 1000 may be a part of a web cloud with the input image of the oral cavity and the report uploaded to the web cloud. The system 1000 involving computer-based process of supervised learning using the convolutional network as described can thus be effectively used to screen the image of the oral cavities. The system 1000 identifies indicators which are further processed to automatically provide indications of relevant pre-cancerous or cancerous tissue, in particular indications of OC. The system 1000 increases efficiency by the utilization of the well trained convolutional network for detecting and classifying the pre-cancerous or cancerous tissues thus providing cost-effective early screening and treatment to the patient.
[0060] The system 1000 reduces the time-consumption involved in a manual process requiring a trained medical practitioner to evaluate digital images of the oral cavity. The system 1000 using the convolutional system 1000 effectively improves the quality of analysis of the image of the oral cavity by detecting indicators of minute size which are often difficult to detect in the manual process of evaluating the image of the oral cavity.

[0061] The present invention described above, although described functionally or sensibly, may be configured to work in a network environment comprising a computer in communication with one or more devices. The present invention, may be implemented by computer programmable instructions stored on one or more computer readable media and executed by a processing unit 102 of the computer. The computer comprises the processing unit 102, a memory unit 103, an input/output (I/O) controller, and a display 104 communicating via a data bus. The computer may comprise multiple processing unit 102s to increase a computing capability of the computer. The processing unit 102 is an electronic circuit which executes computer programs.

[0062] The memory unit 103, for example, comprises a read only memory (ROM) and a random access memory (RAM). The memory unit 103 stores the instructions for execution by the processing unit 102. In this invention, the one or more memory units 103 is the memory unit 103. For instance, the memory unit 103 stores the reference dataset and the reference ground- truth file. The memory unit 103 may also store intermediate, static and temporary information required by the processing unit 102 during the execution of the instructions. The computer comprises one or more input devices, for example, a keyboard such as an alphanumeric keyboard, a mouse, a joystick, etc. The I/O controller controls the input and output actions performed by a user. The data bus allows communication between modules of the computer. The computer directly or indirectly communicates with the devices via an interface, for example, a local area network (LAN), a wide area network (WAN) or the Ethernet, the Internet, a token ring, or the like. Further, each of the devices adapted to communicate with the computer and may comprise computers with, for example, Sun® processing unit 102s 102, IBM® processing unit 102s 102, Intel® processing unit 102s 102, AMD® processing unit 102s 102, etc.

[0063] The computer readable media comprises, for example, CDs, DVDs, floppy disks, optical disks, magnetic-optical disks, ROMs, RAMs, EEPROMs, magnetic cards, application specific integrated circuits (ASICs), or the like. Each of the computer readable media is coupled to the data bus.

[0064] The foregoing examples have been provided merely for the purpose of explanation and does not limit the present invention disclosed herein. While the invention has been described with reference to various embodiments, it is understood that the words are used for illustration and are not limiting. Those skilled in the art, may effect numerous modifications thereto and changes may be made without departing from the scope and spirit of the invention in its aspects.

[0065] In an embodiment, the imaging device 101 employed for screening the oral cancer may be covered with a disposable sheath that fits over the distal end of the mobile phone. The use of a disposable sheath during the screening of a patient's oral cavity for precancerous and/or cancerous tissue protects the screening device from coming into intimate contact with the patient. Thus, after screening the oral cavity of a first patient with the screening device, the first disposable sheath can be properly disposed of as biological waste material. A new disposable sheath can then be placed on the screening device and used to perform an oral cavity screening of a second patient without having to disinfect the entire device.
 
, Claims:What is claimed is:
1. A device for determining a patient risk for oral cancer, the device comprising:
an imaging device 101 configured for capturing at least one image of the oral cavity of a subject;
a processing unit 102 coupled to the imaging device 101, the processing unit 102 configured for processing the image for identifying the probable presence of a pre-cancerous or a cancerous tissue and wherein the processing comprises:
a pre-processing the image;
identifying a plurality of indicators throughout the image using a convolutional network;
detecting a primary index depicting probability of occurrence of oral cancer in the subject, based the identified indicators using the convolutional network; and
displaying the primary index in a display unit 104 coupled to the processing unit 102.

2. The system of claim 1, further comprising an illumination unit configured for illuminating the oral cavity of the subject, in addition to the illumination achieved by the image capturing device.

3. The system of claim 1, wherein the imaging device 101 is a portable handheld device with a built in illumination device.

4. A method of screening oral cancer, the method comprising steps of:
capturing at least one image of the oral cavity of a subject, using an imaging device 101;
processing the image for identifying probable presence of a pre-cancerous or a cancerous tissue, by a processing unit 102 and wherein the processing comprises:
pre-processing the image;
identifying a plurality of indicators throughout the image using a convolutional network;
detecting a primary index depicting probability of occurrence of oral cancer in the subject, based the identified indicators using the convolutional network; and
displaying the primary index on a display unit 104 coupled to the processing unit 102.

5. The method of claim 4, wherein the indicator represents a risk score allowing a user to distinguish between a benign lesion, a dysplastic lesion, and a cancerous lesion.

6. A computer implemented method of screening oral cancer, the method comprising:
providing a memory unit 103 and a processing unit 102, the memory unit 103 configured for storing instructions to be executed by the processing unit 102, which when executed cause the processing unit 102 to:
capture at least one image of the oral cavity of a subject, using an imaging device 101;
process the image for identifying probable presence of an oral cancer and wherein the processing comprises:
pre-process the image;
identify a plurality of indicators throughout the image using a convolutional network;
detect a primary index depicting probability of occurrence of oral cancer in the subject, based the identified indicators using the convolutional network; and
display the primary index on a display unit 104 coupled to the processing unit 102.

7. The computer implemented method of claim 6, wherein the indicator represents a risk score allowing a user to distinguish between a benign lesion, a dysplastic lesion, and a cancerous lesion.