scholarly journals Using artificial intelligence in a primary care setting to identify patients at risk for cancer: a risk prediction model based on routine laboratory tests

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Patricia Diana Soerensen ◽  
Henry Christensen ◽  
Soeren Gray Worsoe Laursen ◽  
Christian Hardahl ◽  
Ivan Brandslund ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Primary Care ◽  
Risk Score ◽  
Routine Laboratory ◽  
Operating Characteristics ◽  
Blood Tests ◽  
Validation Test ◽  
Model Based ◽  
Increased Risk ◽  
Risk Of Cancer

Abstract Objectives To evaluate the ability of an artificial intelligence (AI) model to predict the risk of cancer in patients referred from primary care based on routine blood tests. Results obtained with the AI model are compared to results based on logistic regression (LR). Methods An analytical profile consisting of 25 predefined routine laboratory blood tests was introduced to general practitioners (GPs) to be used for patients with non-specific symptoms, as an additional tool to identify individuals at increased risk of cancer. Consecutive analytical profiles ordered by GPs from November 29th 2011 until March 1st 2020 were included. AI and LR analysis were performed on data from 6,592 analytical profiles for their ability to detect cancer. Cohort I for model development included 5,224 analytical profiles ordered by GP’s from November 29th 2011 until the December 31st 2018, while 1,368 analytical profiles included from January 1st 2019 until March 1st 2020 constituted the “out of time” validation test Cohort II. The main outcome measure was a cancer diagnosis within 90 days. Results The AI model based on routine laboratory blood tests can provide an easy-to use risk score to predict cancer within 90 days. Results obtained with the AI model were comparable to results from the LR model. In the internal validation Cohort IB, the AI model provided slightly better results than the LR analysis both in terms of the area under the receiver operating characteristics curve (AUC) and PPV, sensitivity/specificity while in the “out of time” validation test Cohort II, the obtained results were comparable. Conclusions The AI risk score may be a valuable tool in the clinical decision-making. The score should be further validated to determine its applicability in other populations.

scholarly journals Prediction of risk of prolonged post-concussion symptoms: Derivation and validation of the TRICORDRR (Toronto Rehabilitation Institute Concussion Outcome Determination and Rehab Recommendations) score

PLoS Medicine ◽  
2021 ◽  
Vol 18 (7) ◽  
pp. e1003652
Author(s):  
Laura Kathleen Langer ◽  
Seyed Mohammad Alavinia ◽  
David Wyndham Lawrence ◽  
Sarah Elizabeth Patricia Munce ◽  
Alice Kam ◽  
...  
Keyword(s):  
Primary Care ◽  
Logistic Regression ◽  
Cohort Study ◽  
Risk Score ◽  
Demographic Factors ◽  
Limited Information ◽  
Derivation Cohort ◽  
Injury Index ◽  
Increased Risk ◽  
Image Position

Background Approximately 10% to 20% of people with concussion experience prolonged post-concussion symptoms (PPCS). There is limited information identifying risk factors for PPCS in adult populations. This study aimed to derive a risk score for PPCS by determining which demographic factors, premorbid health conditions, and healthcare utilization patterns are associated with need for prolonged concussion care among a large cohort of adults with concussion. Methods and findings Data from a cohort study (Ontario Concussion Cohort study, 2008 to 2016; n = 1,330,336) including all adults with a concussion diagnosis by either primary care physician (ICD-9 code 850) or in emergency department (ICD-10 code S06) and 2 years of healthcare tracking postinjury (2008 to 2014, n = 587,057) were used in a retrospective analysis. Approximately 42.4% of the cohort was female, and adults between 18 and 30 years was the largest age group (31.0%). PPCS was defined as 2 or more specialist visits for concussion-related symptoms more than 6 months after injury index date. Approximately 13% (73,122) of the cohort had PPCS. Total cohort was divided into Derivation (2009 to 2013, n = 417,335) and Validation cohorts (2009 and 2014, n = 169,722) based upon injury index year. Variables selected a priori such as psychiatric disorders, migraines, sleep disorders, demographic factors, and pre-injury healthcare patterns were entered into multivariable logistic regression and CART modeling in the Derivation Cohort to calculate PPCS estimates and forward selection logistic regression model in the Validation Cohort. Variables with the highest probability of PPCS derived in the Derivation Cohort were: Age >61 years (p^ = 0.54), bipolar disorder (p^ = 0.52), high pre-injury primary care visits per year (p^ = 0.46), personality disorders (p^ = 0.45), and anxiety and depression (p^ = 0.33). The area under the curve (AUC) was 0.79 for the derivation model, 0.79 for bootstrap internal validation of the Derivation Cohort, and 0.64 for the Validation model. A limitation of this study was ability to track healthcare usage only to healthcare providers that submit to Ontario Health Insurance Plan (OHIP); thus, some patients seeking treatment for prolonged symptoms may not be captured in this analysis. Conclusions In this study, we observed that premorbid psychiatric conditions, pre-injury health system usage, and older age were associated with increased risk of a prolonged recovery from concussion. This risk score allows clinicians to calculate an individual’s risk of requiring treatment more than 6 months post-concussion.

ABC score: a new risk score that accurately predicts mortality in acute upper and lower gastrointestinal bleeding: an international multicentre study

Gut ◽  
2020 ◽  
pp. gutjnl-2019-320002
Author(s):  
Stig Borbjerg Laursen ◽  
Kathryn Oakland ◽  
Loren Laine ◽  
Vered Bieber ◽  
Riccardo Marmo ◽  
...  
Keyword(s):  
Gastrointestinal Bleeding ◽  
Risk Score ◽  
Early Identification ◽  
Mortality Rates ◽  
Lower Gastrointestinal Bleeding ◽  
Low Risk ◽  
Regression Analyses ◽  
Operating Characteristics ◽  
Risk Of Death ◽  
Blood Tests

ObjectivesExisting scores are not accurate at predicting mortality in upper (UGIB) and lower (LGIB) gastrointestinal bleeding. We aimed to develop and validate a new pre-endoscopy score for predicting mortality in both UGIB and LGIB.Design and settingInternational cohort study. Patients presenting to hospital with UGIB at six international centres were used to develop a risk score for predicting mortality using regression analyses. The score’s performance in UGIB and LGIB was externally validated and compared with existing scores using four international datasets. We calculated areas under receiver operating characteristics curves (AUROCs), sensitivities, specificities and outcome among patients classified as low risk and high risk.Participants and resultsWe included 3012 UGIB patients in the development cohort, and 4019 UGIB and 2336 LGIB patients in the validation cohorts. Age, Blood tests and Comorbidities (ABC) score was closer associated with mortality in UGIB and LGIB (AUROCs: 0.81–84) than existing scores (AUROCs: 0.65–0.75; p≤0.02). In UGIB, patients with low ABC score (≤3), medium ABC score (4–7) and high ABC score (≥8) had 30-day mortality rates of 1.0%, 7.0% and 25%, respectively. Patients classified low risk using ABC score had lower mortality than those classified low risk with AIMS65 (threshold ≤1) (1.0 vs 4.5%; p<0.001). In LGIB, patients with low, medium and high ABC scores had in-hospital mortality rates of 0.6%, 6.3% and 18%, respectively.ConclusionsIn contrast to previous scores, ABC score has good performance for predicting mortality in both UGIB and LGIB, allowing early identification and targeted management of patients at high or low risk of death.

602-P: Explaining an Artificial Intelligence (AI) System for Diabetic Retinopathy (DR) Screening in Primary Care

Diabetes ◽  
2019 ◽  
Vol 68 (Supplement 1) ◽  
pp. 602-P
Author(s):  
NISHIT UMESH PAREKH ◽  
MALAVIKA BHASKARANAND ◽  
CHAITHANYA RAMACHANDRA ◽  
SANDEEP BHAT ◽  
KAUSHAL SOLANKI
Keyword(s):  
Artificial Intelligence ◽  
Primary Care ◽  
Diabetic Retinopathy

Neuroradiology Manifestations of Li-Fraumeni Syndrome: Epidemiology, Genetics, Imaging Findings, and Management

Neurographics ◽  
2020 ◽  
Vol 10 (4) ◽  
pp. 228-235
Author(s):  
S. Naganawa ◽  
T. Donohue ◽  
A. Capizzano ◽  
Y. Ota ◽  
J. Kim ◽  
...  
Keyword(s):  
Cell Cycle ◽  
Tumor Suppressor ◽  
Tumor Suppressor Gene ◽  
Soft Tissue Sarcomas ◽  
Suppressor Gene ◽  
Imaging Findings ◽  
Li Fraumeni Syndrome ◽  
Increased Risk ◽  
Li Fraumeni ◽  
Risk Of Cancer

Li-Fraumeni syndrome is a familial cancer predisposition syndrome associated with germline mutation of the tumor suppressor gene 53, which encodes the tumor suppressor p53 protein. Affected patients are predisposed to an increased risk of cancer development, including soft-tissue sarcomas, breast cancer, brain tumors, and adrenocortical carcinoma, among other malignancies. The tumor suppressor gene TP53 plays an important, complex role in regulating the cell cycle, collaborating with transcription factors and other proteins. The disruption of appropriate cell cycle regulation by mutated TP53 is considered to be the cause of tumorigenesis in Li-Fraumeni syndrome. Appropriate surveillance, predominantly by using MR imaging, is used for early malignancy screening in an effort to improve the survival rate among individuals who are affected. Patients with Li-Fraumeni syndrome are also at increased risk for neoplasm development after radiation exposure, and, therefore, avoiding unnecessary radiation in both the diagnostic and therapeutic settings is paramount. Here, we review the epidemiology, genetics, imaging findings, and the current standard surveillance protocol for Li-Fraumeni syndrome from the National Comprehensive Cancer Network as well as potential treatment options.Learning Objective: Describe the cause of second primary malignancy among patients with Li-Fraumeni syndrome.

scholarly journals Occasional and persistent frequent attenders and sickness absences in occupational health primary care: a longitudinal study in Finland

BMJ Open ◽  
2019 ◽  
Vol 9 (2) ◽  
pp. e024980 ◽  
Author(s):  
Tiia T M Reho ◽  
Salla A Atkins ◽  
Nina Talola ◽  
Markku P T Sumanen ◽  
Mervi Viljamaa ◽  
...  
Keyword(s):  
Primary Care ◽  
Longitudinal Study ◽  
Occupational Health ◽  
Sick Leave ◽  
Diagnostic Codes ◽  
Frequent Attenders ◽  
Frequent Attendance ◽  
Increased Risk ◽  
Working Age

ObjectivesFrequent attenders (FAs) create a substantial portion of primary care workload but little is known about FAs’ sickness absences. The aim of the study is to investigate how occasional and persistent frequent attendance is associated with sickness absences among the working population in occupational health (OH) primary care.Setting and participantsThis is a longitudinal study using medical record data (2014–2016) from an OH care provider in Finland. In total, 59 676 patients were included and categorised into occasional and persistent FAs or non-FAs. Sick-leave episodes and their lengths were collected along with associated diagnostic codes. Logistic regression was used to analyse associations between FA status and sick leaves of different lengths (1–3, 4–14 and ≥15 days).ResultsBoth occasional and persistent FA had more and longer duration of sick leave than non-FA through the study years. Persistent FAs had consistently high absence rates. Occasional FAs had elevated absence rates even 2 years after their frequent attendance period. Persistent FAs (OR=11 95% CI 7.54 to 16.06 in 2016) and occasional FAs (OR=2.95 95% CI 2.50 to 3.49 in 2016) were associated with long (≥15 days) sickness absence when compared with non-FAs. Both groups of FAs had an increased risk of long-term sick leaves indicating a risk of disability pension.ConclusionBoth occasional and persistent FAs should be identified in primary care units caring for working-age patients. As frequent attendance is associated with long sickness absences and possibly disability pensions, rehabilitation should be directed at this group to prevent work disability.

scholarly journals Cohort profile: CROSS-TRACKS: a population-based open cohort across healthcare sectors in Denmark

BMJ Open ◽  
2020 ◽  
Vol 10 (10) ◽  
pp. e039996
Author(s):  
Anders Hammerich Riis ◽  
Pia Kjær Kristensen ◽  
Matilde Grøndahl Petersen ◽  
Ninna Hinchely Ebdrup ◽  
Simon Meyer Lauritsen ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Primary Care ◽  
Hospital Admissions ◽  
Smoking Status ◽  
Population Based ◽  
Fertility Treatment ◽  
Study Patient ◽  
Transfer Payment ◽  
Sociodemographic Characteristics ◽  
Patient Pathways

PurposeThis paper describes the open cohort CROSS-TRACKS, which comprises population-based data from primary care, secondary care and national registries to study patient pathways and transitions across sectors while adjusting for sociodemographic characteristics.ParticipantsA total of 221 283 individuals resided in the four Danish municipalities that constituted the catchment area of Horsens Regional Hospital in 2012–2018. A total of 96% of the population used primary care, 35% received at least one transfer payment and 66% was in contact with a hospital at least once in the period. Additional clinical information is available for hospital contacts (eg, alcohol intake, smoking status, body mass index and blood pressure). A total of 23% (n=8191) of individuals aged ≥65 years had at least one potentially preventable hospital admission, and 73% (n=5941) of these individuals had more than one.Findings to dateThe cohort is currently used for research projects in epidemiology and artificial intelligence. These projects comprise a prediction model for potentially preventable hospital admissions, a clinical decision support system based on artificial intelligence, prevention of medication errors in the transition between sectors, health behaviour and sociodemographic characteristics of men and women prior to fertility treatment, and a recently published study applying machine learning methods for early detection of sepsis.Future plansThe CROSS-TRACKS cohort will be expanded to comprise the entire Central Denmark Region consisting of 1.3 million residents. The cohort can provide new knowledge on how to best organise interventions across healthcare sectors and prevent potentially preventable hospital admissions. Such knowledge would benefit both the individual citizen and society as a whole.

Sign in / Sign up

Export Citation Format

Share Document