Patent Application
20080229246
The field of this invention is the area of medical systems, specifically plastic surgeons, dermatologists, and other physicians performing cosmetic procedures or other specialties that use photographic images as an integral part of their practices.
As a matter of routine, plastic surgeons, dermatologists, and other physicians performing cosmetic procedures, take photographs of their patients for patient photographic documentation. This documentation includes: before and after photos, to show results, to share with colleagues, and to prepare for the surgeries they are going to undertake. Plastic surgery residents often photograph most of their patients for educational purposes.
A guide on what photos should be considered and how to take them is published jointly by the American Society of Plastic Surgeons and The Plastic Surgery Educational Foundation and is entitled “Photographic Standards in Plastic Surgery.” It is a series of 12 “templates” for different parts of the body and not only suggests what photos to take, but how they should be taken in terms of distance and framing. This guide has a single model (female) and a suggested number of photos to be taken and the poses for each photo. As one can imagine, at times, a different pose or other photos can be desired.
Even using digital photography, the method of matching the digital photos to the template is tedious and time consuming. Often, application packages for digital editing (like PhotoShop from Adobe) are used to try and match the photos taken to the suggested photos in the guide. In addition, the standard problems of digital photography present themselves as well. These include downloading of the images, getting consistent color (especially from different cameras or different conditions and photos taken at different times (for the before and after photos or subsequent surgeries, for example). Additionally, measurements on the photograph may need to be taken. Storing the images (often in multiple locations and with specific image formats like DICOM) need to be supported as well as collaboration with other clinicians of sharing of information is left to the user as a task that is handled outside of the image manipulations.
Clinicians collect information about the patient as a matter of routine. This information is rarely attached to the images and not often utilized for actions utilizing the images. The workflow that is utilized by the clinicians, both the surgeon and their staff, can be greatly improved by optimizing the process of taking, manipulating, storing and sharing the images in a single application. In this application, clinician is defined as anyone on staff utilizing the present invention. Some templates shown do not have facial images in them as part of the template. By providing a simple means to add this to their process, one can easily see how errors can be reduced.
Prior art in this area includes both analog (non-digital) examples and those that have utilized aspects of digital photography. An example of the color discrepancies that can occur is shown in the Niamtu Imaging Systems website (see below) or in cosmetic surgery texts such as “Surgical Rejuvenation of the Face” by Thomas J Baker, MD and Howard L Gordon, MD (C. V. Cosby Co., 1986) and “Cosmetic Dermatolologic Surgery” Leonard M. Dzubow, MD (Lippencott/Raven, 1998). Software from digital cameras, like Kodak's EasyShare software, allows for images to be downloaded from the cameras relatively simply and store them logically, for example, by date. Kodak's EasyShare Gallery allows images to be uploaded and shared with others, although downloading of full resolution images by others is not allowed.
Templates are used in many software applications, including Professional Photographers and PictueIt from Microsoft. These applications allow for the sizing of images to suit the individual. Automated sizing of multiple photos on a page and optimizing the size of the individual pictures on that page is shown by commonly-assigned copending U.S. patent application Ser. No. 09/559,478, filed Apr. 27, 2000, entitled Method of Organizing Digital Images on a Page, by Richard A. Simon. Algorithms that find faces within a photograph and recognize objects within photographs are well known in the art, especially in consumer and professional photography applications and, more recently, in the Homeland Security area. Synthetic models of humans are shown using software packages such as Poser from e-frontier (www.e-frontier.com).
The workflow that a clinician follows can vary from one person to another, whether it is their standard practice, what their comforts and preferences are, or simply performing different functions within the same office. For this reason, the handling of the workflows in an application package of this nature needs to be flexible enough to handle them.
Canfield (www.canfieldsci.com) is a provider of camera systems and software to the plastic surgeons, dermatologists and other physicians performing cosmetic procedures. Their products range from cameras to camera systems to software specifically designed to take and analyze images for these specialties. They do not, however, assess and optimize the workflow of these clinicians nor are they particularly easy to use. They are relatively complicated cameras and do not address issues such as automated download and storage within the clinician's system, adding the images to a customized template, or any of the template features offered in the present invention. There is a direct analogy to consumer digital cameras, there is software to support the camera, but the bulk of what happens after the download is left to the user to handle. Canfield solutions are expensive and require specialized equipment in an effort to make images reproducible. The present invention requires no specialized equipment.
Color targets (for color consistency and color management) are well known in the art. Examples of companies that provide color targets for this purpose are MacBeth and Eastman Kodak Company. Photogrammetry (the ability to make measurements from photographs is also a well known science. The American Society of Photogrammetry and Remote Sensing, Manual of Photogrammetry, 5th edition, 2004 (Chris McGlone—Editor, Published by ASPRS) shows how this is done.
In U.S. Patent Application Publication No. 2002/0092534 A1 (Shamoun) a networked system for previewing potential effects of cosmetic surgery procedures. The present invention does not predict effect, but concentrates on the workflow aspects of the steps prior to the surgery without any prediction of outcome. While the present invention shows past results of other patients, no effect of the current patient is provided.
Similarly, U.S. Patent Application Publication Nos. 2002/0009214 A1 (Arima), 2002/0064302 A1 (Massengill), and 2005/0203495 A1 (Malak) refer to procedural methods of assisting with the surgery rather than improving the workflow of the steps before the surgery or showing pre-surgical information within the OR, without any predictive outcome methods as shown in these applications.
There are several offerings in the area of cosmetic and plastic reconstructive surgery that mention photographic images and systems within their offerings. These can be found on the Internet and examples include:
Each of these sites either provides a service to make a “before and after” photograph or attempts to predict the results of a surgery on an individual. There is nothing about the improvement of the workflow within a clinician's office nor mentions about improvements in the way the images are taken, edited, stored and/or shared for collaborative purposes. One such site, Profect Medical Systems, offers a photographic system, much like the Canfield offering, but does not assist in the management, manipulation or other aspects mentioned in the present invention. Niamtu Imaging Systems does offer image editing, but only for “before and after” images to attempt to make them look the same in terms of size and lighting. They only attempt to match the original image of the patient to one taken later and make no attempt to match this automatically, only to use standard image editing tools to do this (resize, adjust contrast, brightness, etc.).
The present invention creates a synthetic model to assist in taking the proper photos for many different purposes, not just “before and after” photos; these include: photos taken for use in surgery, teaching purposes, documentation, multiple procedures, training aids, and assistance is allowing non-clinical personnel to perform the photographic taking and editing in accordance to pre-determined needs.
Medical Modeling is a site that allows models to be created for use in medical applications. This site can be used as a source of the models used in the present invention in the same way Poser from e-frontier can be used. It does not, however, offer the workflow or the automation of that workflow seen in the present invention, nor does it provide for customized templates showing the photos that are to be taken for the purposes stated above.
Briefly, according to one aspect of the present invention a system for plastic surgery comprises entering patient information into a database; computing a template for the patient based on the information; and inserting a synthetic model into the template.
The present invention allows for a camera agnostic methodology for clinicians to easily bring in photographs into an application specifically designed to optimize their workflow, minimize the manipulation of images, allow for data to be added to the images, advanced storage and retrieval capabilities, and allow for automated collaboration and usage in other applications.
It is a software application with optional storage hardware and utilizes customizable menus and preferences on data, searching and modifying templates for images. Instead of using a human model, a synthetic model is used. The model used is determined by the data for the particular patient. This data entry is part of the application.
The templates used are completely modifiable so that other or additional cell images can be added to, or substituted for, in a template. The software allows for alignment lines to be added to the cells. This allows for different poses to align themselves with each other along a common point.
a is a flowchart of typical workflow for plastic surgery clinicians.
b shows modification of the default workflow.
a is an example of a color/measurement target.
b illustrates how measurements can be taken with the target in the photo.
a and 8b show template modification screens.
The present invention has specifics in cosmetic and plastic surgery but can be used in other specialties where photographs are an integral part of the data collection process. This includes dermatology, dentistry, and others. It combines aspects of separate systems, allows for customization of the workflow within an office (even to different clinicians within an office), allows for manual tasks to be done automatically and combines image and patient data with multiple storage options and sharing capabilities. For the purpose of the present invention, workflow is defined as “A process description of how tasks are done, by whom, in what order and how quickly. Workflow can be used in the context of electronic systems or people, i.e. an electronic workflow system can help automate a physician's personal workflow.” The source of this definition is “Healthcare Informatics Online” and the URL is:
http://www.theebusinesssite.com/IT%20Terms/Health%20Terms.htm#sectW
In order to understand the present invention, one needs to understand the workflow in a typical cosmetic surgeon's (and others previously mentioned) office. In this scenario, the clinician can be the doctor, nurse, or a trained assistant. In fact, it may be a different person at specific steps.
a is a flowchart that shows a typical example of pre-surgical workflow for a plastic surgeon. The first step is a meeting between the patient and the clinician(s) 110 to discuss the patient problem and talk about the procedural alternatives that are to be considered. Once it is decided (by both the patient and clinician) that there is something to be done for the patient 120, information about the patient that is pertinent to the case is collected 130. Samples of previous procedures, often called “before and after” photos are shown to the patient 140 so they can get an idea of the results that may be seen in their case. Once a decision of performing the procedure has been reached by the patient 145, the clinician then reviews the photographic standard templates to determine which photos are to be taken 150. Certain situation may occur when the clinician wishes to use a special template or photos that are different 160 than the ones suggested by the template. Photos are then taken 170 of the patient as suggested in the standard template. The software from the camera is typically used to download the images to the computer 180 or a standard interface such as TWAIN is used to bring them into a specific application. The photos are then edited 185 in a application program like PhotoShop (see www.adobe.com) or PaintShop Pro (see www.jasc.com). Typically, zooming, cropping, color adjustments and alignment from picture to picture within a template is done manually with this software. Additionally, the images are then combined into a single image and finally, the images are stored 190 for further use later on.
One can easily see how parts of this workflow would need to be modified for different clinicians and different patients. For example, one may choose to show the “before and after” images 140 before data on the patient is collected 130 or one may choose to take the photos 160 before selecting the template 150. Since there are a limited number of templates, a clinician may become familiar with the pictures that need to be taken and not need to reference the template. The present invention will allow for these changes in workflow by allowing a dynamic menuing structure that can be easily modified. This is shown in
In dealing with photographs, especially those taken at different times and different conditions (lighting, backgrounds, different cameras, etc), it can be difficult to control the color of the images. Color differences can have significant meaning in dealing with medical images and a means to allow consistent color is important to the clinicians. In addition, there are times when it would be desirable to make measurements on the photographs (the science is known as Photogrammetry).
There is a need to define some terms for the present invention. A template is defined as a set of pictures designed to suggest the pictures to be taken for procedure on a particular part of the body.
All of the data fields shown in
One example is customization of the model that is used for overlaying patient photos.
An example of how a synthetic model is advantaged over a human model is illustrated in the case of a very large male patient about to undergo a series of procedures to sculpt his body via liposuction and body sculpting surgeries. It is very cumbersome to try and match the patient images (different height, weight, sex, body type, etc.) to the slender female in the template, as well as set up the alignment lines. A synthetic model of the approximate weight, height and sex of the patient with the same body type would make this very simple. Software such as Poser from e-frontier allows these synthetic models to be generated. This can be done on the fly with the data provided or a set of models can be pre-rendered. Examples of these poser models are abundant on the Internet.
Note that the synthetic model in 620 is in its basest form and features such as hair and clothing can easily be added in software applications like the aforementioned Poser software. In this example, patient information like gender, age, weight and body mass index can be used to find a pre-rendered model that most closely approximates the patient. Additionally, in another embodiment, the same characteristics can be used to generate a model directly from the software that generates the model and completely customized to the particular patient. There are other advantages to using a synthetic model over a human one, including the time and cost to employ a human model and licensing and royalty fees that can incur. In addition, the model is separable from the background and is a distinct object that can be scaled, moved or posed within each cell of the template. If desired, the model can even be made to look like the patient by mapping the patient's photograph onto the model (well-know in the art of photography and 3D-modeling). Software like Poser allows modification of almost every part of the body. Examples are a male emaciated body 630 and male with a heavy body 640 or a body with a heavy torso and normal lower body 650. These synthetic models can be exported to 3D packages that would allow further functionality to be implemented. It is also possible with current know technology to be able to automatically map photos onto these models. Technology examples include, but are not limited to, face finding so that we can automatically place a patient image into a template cell of a face and object recognition technology that can identify a body part (torso, hand, foot, finger, etc) and automatically place patient photos into these templates. In addition, Poser provides for the models to be edited so that information for a particular patient can be used to provide a reasonable model for each individual.
a and 8b illustrates another workflow improvement over current methodologies. In this case, the clinician is allowed to modify a template for a particular procedure and replace and/or remove any of the cells within a template.
Once a template has been chosen, the present invention allows for a modification option 800. Within the templates main area 805, there is a tab or selection for modifying the template chosen 810. A method is shown on how to add 830 or delete 840 a cell from the template. If a different number of cells (from the original template) are used, the template will automatically resize and realign the cells to optimize placement on the page. This is done using a means shown in commonly-assigned copending U.S. patent application Ser. No. 09/559,478, filed Apr. 27, 2000, entitled Method of Organizing Digital Images on a Page, by Richard A. Simon. Taking this a step further, it can be seen how a photo can be taken of a patient and used in several different templates by simply cropping and zooming the photo appropriately. A photo can be taken of the entire body and be used for the facial templates, mid, and lower body templates by zooming in and cropping the image. With digital cameras routinely having the ability to take 5-20 Mega pixel photos, the resolution is more than enough to make this possible.
In this example of modifying a template, it is desired to remove the middle cell 820 which is a ¾ profile and replace it with a left profile 860. This cell is chosen from a library of poses and templates 855 that were pre-rendered for this purpose. If desired, a 3D model can be used and made to move into any position and pose desired. While this may provide more functionality, the time taken to do this could be a productivity problem. In the preferred embodiment, this is an option, but not the standard means of providing new cells for modification. Once the new template has been created, it can be saved in the library 870 for later use, saved in a patient library for use with a particular patient only 875, or can replace the default template 880 within the standard template area 850.
While this functionality works with a human model and taking photos of the model with different pose changes, it is much more cost effective using the synthetic model. Not only will the model not be required for shots that were not taken (cost and time advantages), but specific model modifications are possible with the synthetic version (hair, facial feature modifications, etc). Specific features of a patient can automatically be detected and applied to the model directly that would enhance the ease of photo placement. Examples are facial shape, eye parameters, lip and nose size and shape, and many others. Advancements in face-finding algorithms and object recognition make this a reasonable feature, as long as the workflow is not interrupted or extended. This capability enables any body type, and any pose of any part of the body (as well as the entire body). This flexibility greatly enhances the workflow and customization of the processes involved in this type of application. Since software like Poser allows for animations to occur as well, a model can be animated to determine the pose in any particular patient case.
The workflow now moves to the Import functionality 910. Images are selected using standard OS methods (explorer, “open”, or camera and scan directly into the application using TWAIN or similar methods) and brought together with the chosen template onto a placement screen 900. The current art has the clinician using a different, general purpose application to create the template images (PhotoShop, PaintShop Pro). This is a painstaking process that requires skill in the use of these applications and the applications are not set up to perform the specific functions as the current invention. Observations on clinical workflow have seen as much as 30 minutes to perform this task when it can be done in less than a minute with the current invention. The appropriate photo is chosen from the thumbnails 920 and placed into the appropriate cell in the template where the image is aligned and sized to the model in that cell. This function can be automated where the proper image for the cell is automatically selected (via image analysis looking for a particular pose and features), placed within the proper cell, and sized properly (using face detection and facial feature finding on both the cell model and the patient photo) and placed properly within the cell. All of the technologies mentioned here are well known in the art of professional photography. A comment area for clinician notes 930 is also provided.
Several features are shown to aid in the placement of these images into the cells by the clinician. An outline view 935 allows only an outline of the synthetic model to be seen (as opposed to the fully rendered model). It has been observed that some clinicians find on outline easier than an overlay on a fully rendered model. Another feature is alignment from photo to photo within a template. This is recommended and shown on the physical brochure showing the templates. The alignment feature 940 allows lines (across the cells within the template) to be added that shows alignment to a common feature or features (nose, ears, hips, etc.). The user can add as many of these alignment lines as desired in the X or Y dimension (horizontal and vertical). The model within the cells can also be moved (X and Y) within the cell, as well as the lines themselves, to allow for different type of alignments.
Opacity is the degree of visibility of the template and the photo so that they can be overlaid and matched. The opacity feature 950 provides an interactive means to control how opaque the photo or the template is when matched. Fine tuning of the image to the template may be desired, especially around body extremities. The fine tuning feature 960 allow any of the cells to be seen full screen and zoomed to a finer level.
An additional feature of the current invention is the ID photo embedded into the application. The concept of an ID photo associated with a patient record is not new. This feature simply allows for the integration of that at the same time photos are used for another purpose (placing them into templates). This is another workflow improvement. There is no longer the need to do this as an independent function using another piece of software. The ID photo can be of significant importance in reducing clinical errors. One of the key outputs of the current invention is for use in the operating room (OR) as a key to the surgeon as to what needs to be done. Many of the templates do not have the patient's face in them. With this, the photo is always available to the surgeon as another patient check. In the current invention, a photo of the patient's face is dragged into the ID photo icon 970 and this is kept s part of the template and file.
There are significant workflow gains to be realized when the effort to construct the templates is completed and the clinician proceeds to next steps. There are several ways in which these finalized templates are used and shared.
The standard save for use within the application 1010 allows for the clinician to stop the work short of completion and continue at a later time. Saving the work as an image file 1020 allows for the image to be used in other applications that accept standard image files (JPEG, BMP, etc.). The option to save the individual image cells 1030 allows for a single, or selected multiple images, to be saved in a standard image format. A “clipboard” save is a standard Microsoft Windows feature for quick pasting into other applications. This is shown as the clipboard button 1040. The entire file (images, metadata, and links to the files) can be saved to a CD 1050 for use in an off-site area, such as an OR. commonly-assigned copending U.S. patent application Ser. No. 11/555,313, filed Nov. 1, 2006, entitled Automated Custom Report Generation System for Medical Information, by Squilla et al. shows an example of such an offsite application where this information can be incorporated. By having a CD (or other portable storage, like a jump drive), the clinician is able to bring the data without the dependency on a network or the Internet. This can be especially useful in secure settings or where computer access is limited. The clinician can also provide their computer, if desired. Each, all, or any combination of these “save” options is selectable. When a choice 1010, 1020, 1030, 1040 or 1050 is made, the selection stays highlighted until it is selected again, when that choice is turned off. The same is true for the “share” options 1060. In this case, the options allow for an e-mail 1070, collaboration 1080 or other sharing capabilities (video conferencing, net meetings, etc.). Linking in e-mails is a standard function seen in many Windows applications and technologies such as JPEG and Zoomify allow for high-resolution, high-speed communications of images. As in the “save” menu, these can also be selected at the same time.
The invention has been described in detail with particular reference to certain preferred embodiments thereof, but it will be understood that variations and modifications can be effected within the scope of the invention.
Reference is made to commonly-assigned copending U.S. patent application Ser. No. 11/555,313, filed Nov. 1, 2006, entitled AUTOMATED CUSTOM REPORT GENERATION SYSTEM FOR MEDICAL INFORMATION, by Squilla et al., the disclosure of which is incorporated herein.
