News | Artificial Intelligence | September 10, 2020

Artificial Intelligence Helps Cut Down on MRI No-shows

AJR researchers use modest data, basic engineering AI to solve complex multifactorial operational problem: outpatient MRI appointment no-shows

Six months after deployment, the no-show rate of the predictive model was 15.9%, compared with 19.3% in the preceding 12-month preintervention period — corresponding to a 17.2% improvement from the baseline no-show rate (p < 0.0001). The no-show rates of contactable and noncontactable patients in the group at high risk of appointment no-shows as predicted by the model were 17.5% and 40.3%, respectively (p < 0.0001).

Weekly outpatient MRI appointment no-show rates for 1 year before (19.3%) and 6 months after (15.9%) implementation of intervention measures in March 2019, as guided by XGBoost prediction model. Squares denote data points. Courtesy of the  American Roentgen Ray Society (ARRS), American Journal of Roentgenology (AJR)

September 10, 2020 — According to ARRS’ American Journal of Roentgenology (AJR), artificial intelligence (AI) predictive analytics performed moderately well in solving complex multifactorial operational problems — outpatient MRI appointment no-shows, especially — using a modest amount of data and basic feature engineering.

“Such data may be readily retrievable from frontline information technology systems commonly used in most hospital radiology departments, and they can be readily incorporated into routine workflow practice to improve the efficiency and quality of health care delivery,” wrote lead author Le Roy Chong of Singapore’s Changi General Hospital.

To train and validate their model, Chong and colleagues extracted records of 32,957 outpatient MRI appointments scheduled between January 2016 and December 2018 from their institution’s radiology information system, while acquiring a further holdout test set of 1,080 records from January 2019. Overall, the no-show rate was 17.4%.

After evaluating various machine learning predictive models developed with widely used open-source software tools, Chong and team deployed a decision tree-based ensemble algorithm that uses a gradient boosting framework: XGBoost, version 0.80 [Tianqi Chen].

As Chong et al. explained, “the simple intervention measure of using telephone call reminders for patients with the top 25% highest risk of an appointment no-show as predicted by the model was implemented over 6 months.”

Six months after deployment, the no-show rate of the predictive model was 15.9%, compared with 19.3% in the preceding 12-month preintervention period — corresponding to a 17.2% improvement from the baseline no-show rate (p < 0.0001). The no-show rates of contactable and noncontactable patients in the group at high risk of appointment no-shows as predicted by the model were 17.5% and 40.3%, respectively (p < 0.0001).

Weekly outpatient MRI appointment no-show rates for 1 year before (19.3%) and 6 months after (15.9%) implementation of intervention measures in March 2019, as guided by XGBoost prediction model. Squares denote data points.

“We believe that the main strength of the present study lies in its empirical approach, given the lack of published literature quantifying the impact of actual workflow implementation, with previous studies postulating the potential benefits of applying machine learning techniques to this problem,” the authors of this AJR article concluded. “The aim of our study was not to produce a highly complex model but, rather, to produce one that could be developed relatively quickly, would require minimal data processing, and would be readily deployable in workflow practice for quality improvement.”

For more information: www.arrs.org

Related Content

3D aMRI not only provides a stunning look inside the "beating brain", but it can also measure this physiological motion in all directions. Here, the amplitude of brain motion is overlayed for each brain slice and orientation in 3D. Image credit: 3D aMRI method outlined in Abderezaei et al. Brain Multiphysics (2021); Terem et al. Magnetic Resonance in Medicine (2021).

3D aMRI not only provides a stunning look inside the "beating brain", but it can also measure this physiological motion in all directions. Here, the amplitude of brain motion is overlayed for each brain slice and orientation in 3D. Image credit: 3D aMRI method outlined in Abderezaei et al. Brain Multiphysics (2021); Terem et al. Magnetic Resonance in Medicine (2021).

News | Magnetic Resonance Imaging (MRI) | May 06, 2021
May 6, 2021 — Magnetic Resonance Imaging
Research finds that a commonly used risk-prediction model for lung cancer does not accurately identify high-risk Black patients who could benefit from early screening

Getty Images

News | Lung Imaging | May 05, 2021
May 5, 2021 — Lung cancer is the third most common cance
After radiosurgery concurrent with nivolumab in 59-year-old patient with melanoma BM (patient 1; Supplemental Tables 3 and 5), F-18 FET PET at follow-up 12 weeks after treatment initiation (bottom row) shows significant decrease of metabolic activity (TBRmean, ?28%) compared with baseline (top row), although MRI changes were consistent with progression according to iRANO criteria. Reduction of metabolic activity was associated with stable clinical course over 10 mo. CE = contrast-enhanced. Image created by

After radiosurgery concurrent with nivolumab in 59-year-old patient with melanoma BM (patient 1; Supplemental Tables 3 and 5), F-18 FET PET at follow-up 12 weeks after treatment initiation (bottom row) shows significant decrease of metabolic activity (TBRmean, ?28%) compared with baseline (top row), although MRI changes were consistent with progression according to iRANO criteria. Reduction of metabolic activity was associated with stable clinical course over 10 mo. CE = contrast-enhanced. Image created by N. Galldiks et al., Research Center Juelich, Juelich, Germany.

News | PET Imaging | May 05, 2021
May 5, 2021 — For patients with brain metastases, amino acid ...
When working with a vendor on project implementation, it is critical to focus on the actual execution of the implementation

Getty Images

Feature | Radiology Business | May 05, 2021 | By Jef Williams
You have selected a vendor. Congratulations.
Building the right infrastructure today will ensure the needed tools are there tomorrow, whatever the challenge may be

Getty Images

Feature | Radiology Business | May 05, 2021 | By Tom Cheesewrite in collaboration with Ludger Philippsen
The COVID-19 pandemic came as a shock, but not a
#DDR allows #clinicians to observe movement like never before. This enhanced version of a standard digital radiographic system can acquire up to 15 sequential #radiographs per second resulting in 20 seconds of motion and multiple individual #radiographic images. #DDR is not fluoroscopy; it is #cineradiography, or #Xray that moves.

DDR allows clinicians to observe movement like never before. This enhanced version of a standard digital radiographic system can acquire up to 15 sequential radiographs per second resulting in 20 seconds of motion and multiple individual radiographic images. DDR is not fluoroscopy; it is cineradiography, or X-ray that moves. The resulting images provide clinicians with a 4-D data set (a video) that depicts physiological movement. 

Sponsored Content | Case Study | Digital Radiography (DR) | May 05, 2021
Musculoskeletal injuries can be difficult to diagnose with a traditional X-ray because X-rays only reveal a static im
With the growing adoption of #data #analytics in #healthcare, we are seeing more clearly that there are two sides of data
Feature | Artificial Intelligence | May 05, 2021 | By Sundararajan Mani
With the growing adoption of...