scholarly journals The Likelihood Difference Heuristic and Binary Test Selection Given Situation-Specific Utilities

2021 ◽  
Author(s):  
Jonathan D. Nelson ◽  
Christine Rosenauer ◽  
Vincenzo Crupi ◽  
Katya Tentori ◽  
Björn Meder
Keyword(s):  
Prior Probability ◽  
Information Gain ◽  
Medical Condition ◽  
Real Life ◽  
Latent Tuberculosis ◽  
Medical Test ◽  
False Positive Test ◽  
Other Information ◽  
Test Result ◽  
Therapeutic Threshold

Consider the task of selecting a medical test to determine whether a patient has a particular disease. Normatively, this requires taking into account (i) the prior probability of the disease, (ii) the likelihood---for each available test---of obtaining a positive result if the medical condition is present or absent, respectively, and (iii) the utilities for both correct and incorrect treatment decisions based upon each possible test result. But these quantities may not be precisely known. Are there strategies that could help identify the test with the highest utility given incomplete information? Here we consider the Likelihood Difference Heuristic (LDH), a simple heuristic that selects the test with the highest difference between the likelihood of obtaining a true positive and a false positive test result, ignoring all other information. We prove that the LDH is optimal when the probability of the disease equals the therapeutic threshold, the probability for which treating the patient and not treating the patient have the same expected utility. By contrast, prominent models of the value of information from the literature, such as information gain, probability gain, and Bayesian diagnosticity, are not optimal under these circumstances. Further results show how, depending on the relationship of the therapeutic threshold and prior probability of the disease, it is possible to determine which likelihoods are more important for assessing tests' expected utilities. Finally, to illustrate the potential relevance for real-life contexts, we show how the LDH might be applied to choosing tests for screening of latent tuberculosis infection.

scholarly journals 1391. Latent Tuberculosis Screening Cascade in Liver Transplant Candidates: A Single, Transplant Center Experience

2019 ◽  
Vol 6 (Supplement_2) ◽  
pp. S505-S506
Author(s):  
Natalia Medvedeva ◽  
Harry Cheung ◽  
Elizabeth Wootton ◽  
Kara Ventura ◽  
Marwan M Azar ◽  
...  
Keyword(s):  
Risk Factors ◽  
Liver Transplant ◽  
Screening Method ◽  
Latent Tuberculosis ◽  
Post Transplantation ◽  
Radiographic Findings ◽  
Ltbi Treatment ◽  
Ltbi Screening ◽  
Transplant Candidates ◽  
Test Result

Abstract Background Screening for latent tuberculosis infection (LTBI) is an essential component of the pre-transplant evaluation and key in identifying patients at risk for TB reactivation post-transplantation. At our center, liver transplant candidates (LTC) are routinely referred to transplant infectious disease (TID) for pre-TID evaluation including LTBI screening. Our aim was to determine the effectiveness of our screening practices and identify barriers to LTBI treatment. Methods We conducted a medical chart review of actively wait-listed LTC as of February 18/2019. Data points collected included: TB risk factors, TID referral and completion of evaluation, intention to screen for LTBI (defined as placing an order), screening completion (with documentation of a test result), screening method (IGRA or PPD), screening test result, radiographic findings, and treatment initiation and completion, if applicable. A positive screen was defined as a positive IGRA or PPD result while a negative screen was defined as a negative result or an indeterminate result with lack of epidemiological risk factors and negative radiographic findings. The proportion of LTC who completed each step in the cascade of care for LTBI was determined. Results Of 102 LTC, 100 met inclusion criteria. Two were excluded due to past LTBI treatment. Of 100 LTC, 95 completed a pre-TID evaluation. For 94 (98.9%), there was intention to screen. Of those intended for screening, 91 (95.8%) successfully completed screening; 6 (6.6%) patients screened positive and 85 (93.4%) screened negative. All 6 LTC who tested positive were recommended for treatment. Five of 6 (83.3%) agreed to treatment, 3/6 (50.0%) started treatment, and all 3 completed treatment. Reasons for non-treatment included: deferral until completion of HCV treatment or hepatologist approval or patient refusal. Treatment regimens included rifampin (n = 1) and isoniazid (n = 2). Conclusion The prevalence of LTBI in our LTC cohort was low. Nonetheless, TID played a role in the successful completion of LTBI screening and identifying those appropriate for treatment in this vulnerable patient population. Barriers to successful LTBI screening and treatment completion are contingent on effective care coordination and addressing competing co-morbidities. Disclosures All authors: No reported disclosures.

scholarly journals 789. Comparison of Interferon-γ Release Assays (IGRAs) for Diagnosis of Latent or Active Tuberculosis in Cancer Patients

2018 ◽  
Vol 5 (suppl_1) ◽  
pp. S283-S283
Author(s):  
Prathit Kulkarni ◽  
Joumana Kmeid ◽  
Marjorie Batista ◽  
Firas El Chaer ◽  
Ella Ariza-Heredia ◽  
...  
Keyword(s):  
Cancer Patients ◽  
Hematopoietic Cell Transplantation ◽  
Latent Tuberculosis ◽  
Absolute Lymphocyte Count ◽  
Interferon Γ ◽  
Active Tuberculosis ◽  
Increased Risk ◽  
Group A ◽  
Test Result ◽  
Research Grant

Abstract Background Patients with certain types of cancer are at increased risk for progression from latent tuberculosis infection (LTBI) to active tuberculosis (ATB) because of immunosuppression. The purpose of this study was to compare the utility of the two commonly used IGRAs, QuantiFERON-TB Gold® (QFT) and T-spot.TB® (T-spot.TB), for diagnosis of LTBI or ATB in cancer patients. Methods We identified patients who had an initial IGRA during 2013 and 2014 at our institution. Along with demographic information, collected clinical data included type of underlying cancer or other condition, reason for testing, diagnosis of ATB following testing, and absolute lymphocyte count (ALC) at the time of testing. IGRA results (positive, negative, borderline, or indeterminate/invalid) were compared between patients who underwent testing with either QFT or T-spot.TB. Results A total of 356 patients had 411 QFT tests done, while 737 patients had 853 T-spot.TB tests performed. The most common underlying malignancies in the QFT and T-spot.TB groups were acute myeloid leukemia (30% and 25%, respectively) and solid tumors (28% vs. 30%, respectively). The most common reasons for testing were pre-hematopoietic-cell transplantation (HCT) screening (42% with QFT and 31% with T spot.TB) or suspected pulmonary ATB (34% with QFT and 42% with T spot.TB). In the QFT group, 145/411 (35%) tests were indeterminate, while only 96/853 (11%) tests in the T-spot.TB group were invalid (P <0.001). The median ALC was 650 cells/µL in patients with an indeterminate result in the QFT group and 90 cells/µL in patients with an invalid test in the T-spot.TB group. A total of four patients were diagnosed with ATB at 1 year after testing. Figure 1 provides a flowchart describing IGRA testing results and development of ATB. Conclusion The frequency of an inconclusive test result is significantly higher with QFT as compared with T-spot.TB for diagnosis of LTBI or ATB in cancer patients. A low ALC is likely a contributing factor in indeterminate QFT and invalid T spot.TBresults. Disclosures E. Ariza-Heredia, Oxford Immunotec: Grant Investigator, Research grant. R. F. Chemaly, Oxford Immunotec: Consultant and Grant Investigator, Research grant.

scholarly journals Clinical predictors of a positive test result in patients undergoing genetic evaluation for a hereditary kidney cancer syndrome

2019 ◽  
Vol 14 (8) ◽  
Author(s):  
Andrea Kokorovic ◽  
Aidan Thomas ◽  
Jesus Serrano-Lomelin ◽  
Meghan Ferguson ◽  
Ricardo A. Rendon
Keyword(s):  
High Risk ◽  
Kidney Cancer ◽  
Genetic Counselling ◽  
Real Life ◽  
Medical Genetics ◽  
Positive Test ◽  
Positive Test Result ◽  
Clinical Implementation ◽  
Cancer Syndrome ◽  
Test Result

Introduction: Guidelines are available to assist providers in identifying patients with renal cell carcinoma (RCC) that may benefit from genetic counselling, however, the evidence for these recommendations lacks support from the literature and controversy remains as to who should be referred. We aimed to delineate risk factors associated with a positive genetic test in a real-life cohort of patients with RCC referred to a regional medical genetics unit for evaluation of a hereditary kidney cancer syndrome. Methods: Patients with a diagnosis of RCC referred to Maritime Medical Genetics Service (Nova Scotia, Canada) from 2006–2017 were reviewed using retrospective data. The primary outcome was identification of clinical features that were associated with a positive test result. Logistic regression models were used for analysis. Results: A total of 135 patients were referred to medical genetics for evaluation; 102 patients were evaluated, 75 underwent testing, and 74 were included in the final analysis. Five patients tested positive: three Birt Hogg Dube, one Cowden syndrome, and one Von Hippel Lindau. Presence of dermatological lesions (specifically fibrofolliculomas) and more than two high-risk features were the only predictors of a positive test result. Conclusions: The presence of dermatological lesions and more than two high-risk features are the only predictors of a positive test result in patients with a suspected hereditary kidney cancer syndrome. These findings are not reflected in current guidelines, and the clinical implementation of our results may improve the identification of high-risk patients for genetic counselling.

scholarly journals Discovering hidden information in biosignals from patients using artificial intelligence

2020 ◽  
Vol 73 (4) ◽  
pp. 275-284
Author(s):  
Dukyong Yoon ◽  
Jong-Hwan Jang ◽  
Byung Jin Choi ◽  
Tae Young Kim ◽  
Chang Ho Han
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Deep Learning ◽  
Medical Condition ◽  
Real Life ◽  
Black Box ◽  
Basic Knowledge ◽  
Hidden Information ◽  
Near Future ◽  
Invasive Evaluation

Biosignals such as electrocardiogram or photoplethysmogram are widely used for determining and monitoring the medical condition of patients. It was recently discovered that more information could be gathered from biosignals by applying artificial intelligence (AI). At present, one of the most impactful advancements in AI is deep learning. Deep learning-based models can extract important features from raw data without feature engineering by humans, provided the amount of data is sufficient. This AI-enabled feature presents opportunities to obtain latent information that may be used as a digital biomarker for detecting or predicting a clinical outcome or event without further invasive evaluation. However, the black box model of deep learning is difficult to understand for clinicians familiar with a conventional method of analysis of biosignals. A basic knowledge of AI and machine learning is required for the clinicians to properly interpret the extracted information and to adopt it in clinical practice. This review covers the basics of AI and machine learning, and the feasibility of their application to real-life situations by clinicians in the near future.

scholarly journals An Improvement to a Class of Intensity based Spatial Domain Corner Detection Algorithms using Image Fission and Fusion

2019 ◽  
Vol 9 (2) ◽  
pp. 924-930
Keyword(s):  
Information Gain ◽  
Binary Classification ◽  
Real Life ◽  
Detection Algorithm ◽  
Spatial Domain ◽  
Corner Detection ◽  
Computational Time ◽  
Detection Algorithms ◽  
Bit Plane ◽  
Bit Planes

Corners of an object are important as features for the representation and analysis of its shape in computer vision. Corner detection, particularly in real scenes, is still a challenge. Most of the corner detectors found in the literature generate a number of false corners, which is not acceptable in real-life applications. In this paper, an improvement to a class of corner detection algorithms is presented using image fission/fusion. In this approach, a grayscale image is first divided into several bit-planes. A corner detector is applied on all the bit-planes simultaneously and a threshold (bitplane) is obtained using the concept of information gain. Finally, all the higher bit-plane corners are recombined (up to some thresholded bit-plane) to obtain the final set of corners. Here the corner detection algorithm is considered as a binary classification problem. Experimental results show that this improved approach reduces the number of erroneous corner detection relative to existing spatial domain corner detection algorithms. The improvements are established with the help of a number of performance measures proposed by various researchers. The proposed approach works better with respect to computational time also. This approach can easily be utilized in different low-level image processing applications.

scholarly journals Characteristics of patients with psoriatic arthritis treated with ustekinumab. Evidence from real-life clinical practice

2021 ◽  
pp. 274-283
Author(s):  
Yu. L. Korsakova ◽  
E. Yu. Loginova ◽  
E. E. Gubar ◽  
T. V. Korotaeva
Keyword(s):  
Psoriatic Arthritis ◽  
Clinical Laboratory ◽  
Real Life ◽  
Circulatory System ◽  
Latent Tuberculosis ◽  
Endocrine System ◽  
Gastrointestinal Diseases ◽  
Good Tolerability ◽  
Viral Hepatitis C ◽  
Circulatory System Diseases

Introduction. Psoriatic arthritis (PsA) is a chronic immunoinflammatory disease characterised by involvement of the skin, nail plates, joints, spine and entheses in the inflammatory process. The IL-12/IL-23 inhibitor ustekinumab (UST) is increasingly being used in psoriasis (Ps) and PsA.Aim of the study. To analyze patients with PsA who were under inpatient treatment in the V.A. Nasonova Scientific Research Institute of Rheumatology and Radiology and who were prescribed UST during the period from 2018 to 2020.Material and methods. UST was administered to 17 patients with PsA (9 women and 8 men), mean age was 46.4 ± 11.3 years. Duration of PsA course was 11 ± 10.5 years. Patients underwent clinical, laboratory and instrumental examination, BSA and PASI, DAPSA and BASDAI indices were determined.Results. Patients predominantly had widespread Ps (BSA 18.2 ± 15.9%). Erosive arthritis was present in 94.1% of patients, and sacroiliitis was detected in 100% of patients. PsA activity was high (DAPSA = 44.9 ± 20.9, BASDAI = 6.2 ± 1.5).94% of patients had two or more comorbidities. Circulatory system diseases were observed in 82.4% of patients, liver diseases in 29.5%, gastrointestinal diseases in 47%, endocrine system diseases in 17.6%, viral hepatitis C in 23.5%, latent tuberculosis infection in 17.6%, and joint surgery was performed in 11.2% of patients. The clinical example presented in the article demonstrates good tolerability of UST in a patient with PsA with a number of comorbidities and the possibility to increase the dose of UST from 45 to 90 mg in case of ineffective therapy.Conclusions. The safety profile of UST is good, and it can be administered to patients with cardiovascular diseases, obesity, various infections, including latent tuberculosis, etc. 

Markers of Alzheimer’s disease in handwriting

2022 ◽  
Vol 26 (6) ◽  
pp. 16-28
Author(s):  
Y. G. Chernov ◽  
Zh. A. Zholdasova
Keyword(s):  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Cognitive Impairment ◽  
Pilot Study ◽  
Supportive Therapy ◽  
Warning Signal ◽  
Handwriting Analysis ◽  
Medical Test ◽  
Handwritten Documents ◽  
Test Result

The aim of the research. Alzheimer’s disease is the most common form of dementia. One of the potential tools for early detection of the onset of the disease is the handwriting analysis. It can be a warning signal for a serious medical investigation. The dynamics of handwriting changes are also a good indicator of the progression of the disease and the eff ectiveness of therapy. Methods. The authors have developed two corresponding tests. The fi rst (AD-HS) allows the assessment of handwriting markers of cognitive impairment and Alzheimer’s disease from an available handwriting sample. The second (ADHC) is designed to assess dynamics by comparing two handwritten documents written at diff erent times. Results. The pilot study includes 16 patients who were found to be at diff erent stages of the disease by medical examination. They all provided old handwriting samples dated 10–20 years ago and new handwriting samples specifi cally written as part of the experiment. Evaluation of 36 handwriting characteristics showed that both tests were eff ective in identifying Alzheimer’s disease and its stage. The correlation between the handwriting analysis and the medical test result was 0.62. Conclusion. Further refi nement of the proposed tests and expansion of the research base will enable handwriting exercises to be incorporated into supportive therapy to slow the progression of the disease.

Sign in / Sign up

Export Citation Format

Share Document