Due to the expense and need for extensive patient preparation in a procedure involving anesthesia, many countries don’t use colonoscopies and instead use the less expensive and less-invasive sigmoidoscopy. A sigmoidoscopy is normally performed every five to ten years, with stool samples relatively easy to obtain with tests done annually or biennially.

The study enrolled patients from 2004 through 2011 and was supported by the National Cancer Institute, Indiana University’s Cancer Center, and the Walther Cancer Foundation.

Factors such as gender, age, physical activity levels, lifestyle habits such as smoking and cohabitation history, non-steroidal anti-inflammatory drug use (such as ibuprofen), and other variables are weighed in relation to importance to creating an individual’s personal risk potential.

Two related screening tools were developed:  one applies to patients with limited endoscopy screening for the lower area of the colon through sigmoidoscopy. The other part is directed toward patients with no tests of colon visualization. 

According to Dr. Imperiale, a practicing gastroenterologist who has studied colon cancer risks and screening over the past two decades, the newer composite weighed methods of scoring assessment can, for example, determine whether the risk of a precancerous polyp has certain types of ominous qualities or of colon cancer, making both a reason for the further need for a colonoscopy.  

Breast cancer detection gets a boost from Predictive Analytics

Medical professionals have long known about the problems posed by dense breast tissue when performing a mammogram screening. Dense tissue, more common in younger pre-menopausal women, is harder for a mammogram to image correctly than fatty tissue, leading to missed tumors as well as false positives. Women with dense breast tissue are three to six times more likely to develop breast cancer than women with less-dense tissue, making finding a better way to detect early cancer a critical priority.

Research from the Dense Tissue and Early Breast Neoplasm Screening (DENSE) trial has shown that supplemental MRI screening can be beneficial in detecting early tumors in women with denser tissue, reported lead author Erik Verburg, MSc. 

Most MRIs show normal physiological and anatomical variations not requiring radiological review, with the majority of those screened women not showing suspicious findings with the MRI, he added.

As a result of the trials, researchers set out to create a new method of reducing radiologists’ need for repeated readings. They explored the feasibility of automated triaging methods based on deep learning, using artificial intelligence, and developing algorithms to learn the differences between breast tissue with and without lesions. The deep learning model, which trained with data from seven hospitals and was tested on dense tissue data from eight hospitals, tested over 4,500 MRI data sets. Of the 9,162 breasts tested, 838 had at least one lesion, with 77 malignancies, with another 8,324 showing no lesions.

The deep learning model considered 90.7 percent of MRIs with lesions to be non-normal, flagging them for further radiological examination. Furthermore, it dismissed close to 40 percent as lesion-free, without missing any cancers.

The conclusion was that it was possible to utilize artificial intelligence to dismiss breast cancer MRI screenings without missing any malignancies, with better results than anticipated, while acknowledging that there was plenty of room (60 percent) to improve.

As a result, the AI triaging system could significantly reduce radiologists’ workloads by reducing reading time overall, according to Verburg. In doing so, providers could give more time and attention to more complicated breast MRI results and subsequent examinations.

Currently, the researchers plan to use other datasets to validate the model and eventually deploy it in the next DENSE screening rounds.

CMS: Reimbursement for Predictive Analytics and Related Technologies and Services

According to the latest CMS guidelines, new medical technology or service may be eligible for a new technology add-on payment if estimated costs are incurred with respect to discharges involving the service or technology, the DRG prospective payment rate that would otherwise be applicable to such discharges under the particular subsection is inadequate. As long as the criteria of being new, very costly, and demonstrating a substantial improvement over existing technologies and services are met, such services or technologies are billable.

One challenge is knowing when and how to submit relevant AI claims data to CMS for consideration,

Partner with a company experienced in billing the latest technologies

Cancer and other assessment technologies, including AI and other forms of analytics, will be increasingly utilized into the foreseeable future, making billing these charges potentially challenging for many back-offices and billing departments.

Even experienced billers and coders may have trouble figuring out insurance payers’ and CMS requirements and regulations pertaining to these technologies.

When your practice makes the choice of partnering with an experienced medical billing and practice management service such as M-Scribe, you can be assured that our dedication to keeping up to date on the latest coding, regulatory and technological changes will provide you with the most accurate and complete claims management services in the industry. 

Since 2002, M-Scribe has helped thousands of practices of all sizes and specialties take control over their billing and related RC activities, resulting in faster, complete payment and fewer denials and missed claims. Contact us at 770-666-0470 or email: H.Gibson@m-scribe.com for a complimentary consultation with our consultants to assess your practice’s revenue and management goals and needs to optimize your RCM.

{{cta(‘4e5f7f98-6ee9-47b3-a637-5774f339f5df’,’justifycenter’)}}