A novel index using inflammatory markers improves the diagnosis of hemophagocytic lymphohistiocytosis in patients with hematologic malignancies.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 7563-7563
Author(s):  
Adi Zoref Lorenz ◽  
Jun Murakami ◽  
Liron Hofstetter ◽  
Swaminathan Padmanabhan Iyer ◽  
Ahmed Alotaibi ◽  
...  
Keyword(s):  
Diagnostic Criteria ◽  
Hemophagocytic Lymphohistiocytosis ◽  
Area Under The Curve ◽  
Scoring Systems ◽  
Hematologic Malignancies ◽  
Classification And Regression Tree ◽  
Optimal Cutoff ◽  
Cart Analysis ◽  
Sensitivity Specificity ◽  
Cutoff Levels

7563 Background: Hemophagocytic lymphohistiocytosis (HLH) is a life-threatening inflammatory syndrome that may accompany hematologic malignancies (HM). The diagnosis of HLH in patients with HM (HM-HLH) is confounded by a number of factors: the most commonly used HLH-2004 diagnostic criteria are derived from studies in infants while the Hscore used in adults is not specific for HMs; moreover, most parameters in these scoring systems may reflect features of the underlying HM rather than HLH associated inflammation; and finally specific diagnostic cutoff values for laboratory abnormalities in HM-HLH have not been defined. We therefore conducted a study to optimize the HLH-2004 laboratory thresholds for the diagnosis of HM-HLH. Methods: A multi-center retrospective study in adult patients with HM in whom testing for HLH was performed. HM-HLH was defined as fulfillment of 5/8 HLH-2004 diagnostic criteria. We established the optimal diagnostic cutoff levels for HLH-2004 laboratory parameters using receiver operating curves (ROC) and combined the best performing parameters into a combined index, using binary logistic regression. We then created a clinical decision tree using a Classification and Regression Tree (CART) analysis with all available parameters, using cross validation. We also determined the prognostic value of our combined diagnostic tool. Results: 225 adults were analyzed (112 with HM-HLH per HLH-2004 and 113 with HM only). 35% of patients were evaluated for HLH routinely upon HM diagnosis. Soluble CD25 (sCD25) and ferritin best discriminated HM-HLH from HM, with an area under the curve (AUC) of 0.83 for each. ROC analysis demonstrated an optimal cutoff of > 4190 U/mL for sCD25 (sensitivity/specificity 91%/69%) and an optimal cutoff of > 2636 ng/ml for ferritin (sensitivity/specificity 64%/86%) for HM-HLH. We term the combination of elevated sCD25 and ferritin using optimized cutoff levels the ‘optimized HLH inflammatory’ (OHI) index. This OHI index was highly specific for the diagnosis of HM-HLH (specificity of 92%, sensitivity 79%). CART analysis demonstrated that OHI index positivity was sufficient to diagnose HM-HLH. In patients without a positive OHI index an Hscore > 168 and either splenomegaly or triglycerides > 279 ng/dL can still diagnose HM-HLH. By following this decision pathway, approximately 92% of patients were accurately classified based on HLH-2004. Furthermore, the OHI was better (odds ratio (OR) 7.9; 95% confidence interval (CI) 4.2-14.6) than Hscore >169 (OR 5.5; CI 3.9-9.6) and > 5/8 HLH-2004 (OR 5.3; CI 3-9.3) at predicting mortality at 1 year. Conclusions: The OHI index derived here is a simple tool that can accurately diagnose HLH and predict mortality in patients with hematologic malignancies. Some patients may not need full HLH workup before intervening with therapy that is HLH directed and not only malignancy directed.

scholarly journals A Novel Inflammatory Index Is Sufficient to Identify Hemophagocytic Lymphohistiocytosis in Adult Patients with Hematologic Malignancies

Blood ◽  
2020 ◽  
Vol 136 (Supplement 1) ◽  
pp. 1-2
Author(s):  
Adi Zoref-Lorenz ◽  
Jun Murakami ◽  
Liron Hofstetter ◽  
Swaminathan P Iyer ◽  
Ahmad S. Alotaibi ◽  
...  
Keyword(s):  
Board Of Directors ◽  
Diagnostic Criteria ◽  
Research Funding ◽  
Roc Analysis ◽  
Hematologic Malignancy ◽  
Hematologic Malignancies ◽  
Advisory Committees ◽  
Inflammatory Index ◽  
Sensitivity Specificity ◽  
Cutoff Levels

Introduction: Hemophagocytic lymphohistiocytosis (HLH) is a life-threatening hyper-inflammatory syndrome which may occur in adults with hematologic malignancies (HM). The diagnosis of HLH in this context (HM-HLH) is hindered by a number of factors. First, the currently used HLH 2004 diagnostic criteria are derived from pediatric patients commonly with HLH-associated genetic lesions, a very different population than adults with cancer. Second, most parameters used for diagnosis of HLH are directly impacted by the underlying HM and may reflect the presence of the malignant clone itself rather than an inflammatory process. Finally, appropriate diagnostic cutoff values for laboratory abnormalities in HM-HLH have not been defined. In this study we determine the diagnostic value of the laboratory components of the HLH 2004 diagnostic criteria and establish optimal cutoffs for the diagnosis of HM-HLH in HM patients. Methods: This is a multicenter, retrospective study of adult patients with a hematologic malignancy in whom sCD25 was measured because of clinically suspected HM-HLH or as part of routine screening of patients with a newly diagnosed hematologic malignancy, between January 2012 and March 2020. We considered patients fulfilling the five of eight of the HLH 2004 diagnostic criteria to have HM-HLH. Patients fulfilling fewer than five criteria were assigned to the HM group. These cohorts were well balanced in terms of disease distribution. We established the optimal cutoffs for laboratory parameters used for the diagnosis of HM-HLH using receiver operating curves (ROC) in a discovery cohort and tested their performance in a validation cohort. In order to improve the results obtained using the individual ROC, we then created a combined ROC using parameters demonstrating the highest individual performance (highest area under the curve (AUC)), in order to develop a diagnostic index. Finally, we examined the performance of each parameter in each cohort by using a contingency table and Chi-square and Fisher's exact test to determine the positive predictive value (PPV), negative predictive value (NPV), sensitivity, specificity and likelihood ratio (LR) of disease for each parameter. Results: 212 adults with HM with or without HLH in whom testing for HLH was performed were included in the study. HMs were: B cell lymphoma (41%), T cell lymphoma (26%), Hodgkin lymphoma (9%), acute myeloid leukemia (8%), myelodysplastic syndrome (8%), myeloproliferative neoplasms (5%) and chronic lymphocytic leukemia (4%). 99 (47%) patients had HM-HLH. Despite considerable overlap in laboratory values between the patient groups, all parameters apart from fibrinogen were able to distinguish HM-HLH from HM alone, with ferritin and sCD25 having the greatest discriminatory power. ROC analysis revealed an optimal cutoff value of >5,600 U/mL for sCD25 (sensitivity/specificity 76%/78%, AUC=0.83) and >1,300 ng/ml for ferritin (sensitivity/specificity 76%/76%, AUC=0.83). Combining the two markers to create a novel inflammatory index (HM-INFL) yielded superior diagnostic ability (AUC =0.86). Using HLH 2004 cutoff levels the HM-INFL index had a sensitivity of 94% and NPV of 94% and when using the optimal cutoff levels, it had a specificity of 92% and PPV of 90% (Table 1). Conclusions: HM-INFL is an index comprising only ferritin and sCD25. Using the original HLH 2004 cutoffs the index is an effective screening tool. Using our newly defined cutoff levels obtained by ROC analysis it is highly specific and can be used as a confirmatory test for the diagnosis of HLH in HM patients. These findings also support the hypothesis that HLH in the context of HM is an inflammatory condition associated with immune dysregulation. Disclosures Miller: Foundation Medicines, Inc.: Consultancy. Daver:Daiichi Sankyo: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Bristol-Myers Squibb: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Karyopharm: Research Funding; Servier: Research Funding; Genentech: Research Funding; AbbVie: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Astellas: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novimmune: Research Funding; Gilead: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Amgen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Trovagene: Research Funding; Fate Therapeutics: Research Funding; ImmunoGen: Research Funding; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Jazz: Consultancy, Membership on an entity's Board of Directors or advisory committees; Trillium: Consultancy, Membership on an entity's Board of Directors or advisory committees; Syndax: Consultancy, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Membership on an entity's Board of Directors or advisory committees; KITE: Consultancy, Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees. Jordan:Sobi: Consultancy.

Serum soluble CD25 and ferritin in distinguishing patients with uncomplicated hematologic malignancies from patients with hematologic malignancies complicated by hemophagocytic lymphohistiocytosis.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. e20072-e20072
Author(s):  
Adi Zoref Lorenz ◽  
Liron Hofstetter ◽  
Jun Murakami ◽  
Oren Pasvolsky ◽  
Elad Guber ◽  
...  
Keyword(s):  
Diagnostic Criteria ◽  
Hemophagocytic Lymphohistiocytosis ◽  
Nk Cell ◽  
Serum Levels ◽  
Hematologic Malignancies ◽  
Screening Criteria ◽  
Underlying Malignancy ◽  
Life Threatening ◽  
Sensitivity Specificity ◽  
Inflammatory Syndrome

e20072 Background: Hemophagocytic lymphohistiocytosis (HLH) is a life-threatening hyper-inflammatory syndrome with distinct clinical and laboratory features. In adults, HLH is often associated with underlying malignancy, most commonly hematologic malignancies (HM). HLH occurs in 1% of adults with HM and overall survival can be low as 10-20%. Abnormal serum levels of the inflammatory markers sCD25 and ferritin are diagnostic criteria for familial HLH. However, because these reactants are often elevated in malignancy, appropriate levels for diagnosis in HM-HLH are unknown. In this study, we establish optimal sCD25 and ferritin levels for the diagnosis of HM-HLH in adults. Methods: Patients from three centers in Israel and Japan with HM-HLH and HM in whom sCD25 testing was performed were studied. The diagnosis of HLH was according to the HLH 2004 diagnostic criteria. Initial (at HLH presentation) and the maximum ferritin levels were analyzed. Sensitivity, specificity, and optimal cutoffs were calculated by receiver operating characteristic (ROC). Results: 62 patients with HM's without HLH and 40 patients with HM-HLH were included. The distribution of ages and HM subtypes was similar between groups (mostly B cell, T/NK cell, and Hodgkin's lymphoma). The median sCD25 concentration in HM was 1776 U/ml versus 8077 U/ml in HM-HLH. The median initial/ maximum ferritin levels were 190/202 ng/ml for the HM group and 2267/4515 ng/ml for the HM-HLH group. Both sCD25 and ferritin were very sensitive but nonspecific. sCD25 > 2400 U/ml had a sensitivity/specificity of 95%/65%, while initial ferritin > 500 ng/ml had a sensitivity/specificity of 95%/75%. ROC analysis demonstrated optimal confirmatory cutoff values (maximizing specificity) of > 10,056 ng/ml for sCD25 (sensitivity/specificity 47%/95%). While initial ferritin demonstrated a cutoff of > 5231 ng/ml (sensitivity/specificity 22.5%/95%, AUC = 0.88) the maximum ferritin performed better with the cutoff of > 5748 ng/ml (sensitivity/specificity 45%/95%, AUC = 0.92). Conclusions: Our data suggest that the current HLH 2004 criteria of ferritin > 500 ng/ml and sCD25 > 2400 U/ml are effective screening criteria for the complication of HLH in HM patients. sCD25 > 10,000 U/ml and initial ferritin > 5,200 ng/ml are highly specific. Patients with suspected HM- HLH should have serial ferritin testing which increases specificity of this test. Future prospective studies are needed to confirm these findings.

scholarly journals Validation and verification of predictive salivary biomarkers for oral health

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Nagihan Bostanci ◽  
Konstantinos Mitsakakis ◽  
Beral Afacan ◽  
Kai Bao ◽  
Benita Johannsen ◽  
...  
Keyword(s):  
Oral Health ◽  
Predictive Accuracy ◽  
Statistical Tests ◽  
Area Under The Curve ◽  
Regression Tree ◽  
High Sensitivity ◽  
Oral Health Status ◽  
Classification And Regression Tree ◽  
Cart Analysis ◽  
Salivary Biomarkers

AbstractOral health is important not only due to the diseases emerging in the oral cavity but also due to the direct relation to systemic health. Thus, early and accurate characterization of the oral health status is of utmost importance. There are several salivary biomarkers as candidates for gingivitis and periodontitis, which are major oral health threats, affecting the gums. These need to be verified and validated for their potential use as differentiators of health, gingivitis and periodontitis status, before they are translated to chair-side for diagnostics and personalized monitoring. We aimed to measure 10 candidates using high sensitivity ELISAs in a well-controlled cohort of 127 individuals from three groups: periodontitis (60), gingivitis (31) and healthy (36). The statistical approaches included univariate statistical tests, receiver operating characteristic curves (ROC) with the corresponding Area Under the Curve (AUC) and Classification and Regression Tree (CART) analysis. The main outcomes were that the combination of multiple biomarker assays, rather than the use of single ones, can offer a predictive accuracy of > 90% for gingivitis versus health groups; and 100% for periodontitis versus health and periodontitis versus gingivitis groups. Furthermore, ratios of biomarkers MMP-8, MMP-9 and TIMP-1 were also proven to be powerful differentiating values compared to the single biomarkers.

scholarly journals Cognitive and Behavioral Domains That Reliably Differentiate Normal Aging and Dementia in Down Syndrome

2021 ◽  
Vol 11 (9) ◽  
pp. 1128
Author(s):  
Jordan P. Harp ◽  
Lisa M. Koehl ◽  
Kathryn L. Van Pelt ◽  
Christy L. Hom ◽  
Eric Doran ◽  
...  
Keyword(s):  
Primary Care ◽  
Down Syndrome ◽  
Area Under The Curve ◽  
Regression Tree ◽  
Classification And Regression Tree ◽  
Vineland Adaptive Behavior Scales ◽  
Cart Analysis ◽  
Dementia Screening ◽  
Significant Difference ◽  
Unit Increase

Primary care integration of Down syndrome (DS)-specific dementia screening is strongly advised. The current study employed principal components analysis (PCA) and classification and regression tree (CART) analyses to identify an abbreviated battery for dementia classification. Scale- and subscale-level scores from 141 participants (no dementia n = 68; probable Alzheimer’s disease n = 73), for the Severe Impairment Battery (SIB), Dementia Scale for People with Learning Disabilities (DLD), and Vineland Adaptive Behavior Scales—Second Edition (Vineland-II) were analyzed. Two principle components (PC1, PC2) were identified with the odds of a probable dementia diagnosis increasing 2.54 times per PC1 unit increase and by 3.73 times per PC2 unit increase. CART analysis identified that the DLD sum of cognitive scores (SCS < 35 raw) and Vineland-II community subdomain (<36 raw) scores best classified dementia. No significant difference in the PCA versus CART area under the curve (AUC) was noted (D(65.196) = −0.57683; p = 0.57; PCA AUC = 0.87; CART AUC = 0.91). The PCA sensitivity was 80% and specificity was 70%; CART was 100% and specificity was 81%. These results support an abbreviated dementia screening battery to identify at-risk individuals with DS in primary care settings to guide specialized diagnostic referral.

An improved index for diagnosis and mortality prediction in malignancy associated hemophagocytic lymphohistiocytosis

Blood ◽  
2021 ◽  
Author(s):  
Adi Zoref-Lorenz ◽  
Jun Murakami ◽  
Liron Hofstetter ◽  
Swaminathan P. Iyer ◽  
Ahmad S. Alotaibi ◽  
...  
Keyword(s):  
High Mortality ◽  
Mortality Risk ◽  
Hemophagocytic Lymphohistiocytosis ◽  
Regression Tree ◽  
Hematologic Malignancies ◽  
Mortality Prediction ◽  
Classification And Regression Tree ◽  
Receiver Operating Curve ◽  
High Mortality Risk ◽  
Inflammatory State

Hemophagocytic lymphohistiocytosis (HLH) is a life-threatening inflammatory syndrome that may complicate hematologic malignancies (HM). The appropriateness of current criteria for diagnosing HLH in the context of HMs is unknown because they were developed for children with familial HLH (HLH-2004) or derived from adult patient cohorts in which HMs were underrepresented (HScore). Moreover, many features of these criteria may directly reflect the underlying HM rather than an abnormal inflammatory state. To improve and potentially simplify HLH diagnosis in patients with HMs, we studied an international cohort of 225 adult patients with various HMs both with and without HLH and for whom HLH-2004 criteria were available. We used classification and regression tree and receiver operating curve analysis to identify the most useful diagnostic and prognostic parameters and optimize laboratory cutoff values. Combined elevation of soluble CD25 (&gt;3,900 U/ml) and ferritin (&gt;1,000 ng/ml) best identified HLH-2004 defining features (sensitivity 84%, specificity 81%). Moreover, this combination, which we term the 'optimized HLH inflammatory' (OHI) index, was highly predictive of mortality (hazard ratio 4.3; confidence interval 3.0-6.2) across diverse HMs. Furthermore, the OHI index identified a large group of patients with high mortality risk that were not defined as having HLH by HLH-2004/HScore. Finally, the OHI demonstrates diagnostic and prognostic value when used for routine surveillance of patients with newly diagnosed HMs as well as those with clinically suspected HLH. Thus, we conclude that the OHI index identifies HM patients with an inflammatory state associated with a high mortality risk and warrants further prospective validation.

External Validation of Different Scoring Systems for Suspected Choledocholithiasis

2019 ◽  
Vol 36 (6) ◽  
pp. 530-538
Author(s):  
Nicolò Tamini ◽  
Davide Paolo Bernasconi ◽  
Luca Gianotti
Keyword(s):  
Operating Characteristic ◽  
Bile Duct Stone ◽  
Characteristic Curve ◽  
External Validation ◽  
Area Under The Curve ◽  
Scoring Systems ◽  
Youden Index ◽  
Predictive Values ◽  
Sensitivity Specificity ◽  
Symptoms And Signs

Aim of the Study: The diagnosis of choledocholithiasis is challenging. Previously published scoring systems designed to calculate the risk of choledocholithiasis were evaluated to appraise the diagnostic performance. Patients and Methods: Data of patients who were admitted between 2013 and 2015 with the following characteristics were retrieved: bile stone-related symptoms and signs, and indication to laparoscopic cholecystectomy. To validate and appraise the performance of the 6 scoring systems, the acknowledged domains of each metrics were applied to the present cohort. Sensitivity, specificity, positive, negative predictive, Youden index, and receiver operating characteristic curve with the area under the curve (AUC) values of the scores were calculated. Results: Two-hundred patients were analyzed. The highest sensitivity and specificity were obtained from the Menezes’ (96.6%) and Telem’s (99.3%) metrics respectively. The Telem’s and Menezes’ scores had the best positive (75.0%) and negative (96.4%) predictive values respectively. The best accuracy, as computed by the Youden index and AUC, was found for the Soltan’s scoring system (0.628 and 0.88, respectively). Conclusion: The available scoring systems are precise only in identifying patients with a negligible risk of common bile duct stone, but overall insufficiently accurate to suggest the routine use in clinical practice.

Clock Drawing Test – screening utility for mild cognitive impairment according to different scoring systems: results of the Leipzig Longitudinal Study of the Aged (LEILA 75+)

2011 ◽  
Vol 23 (10) ◽  
pp. 1592-1601 ◽  
Author(s):  
Lena Ehreke ◽  
Tobias Luck ◽  
Melanie Luppa ◽  
Hans-Helmut König ◽  
Arno Villringer ◽  
...  
Keyword(s):  
Cognitive Impairment ◽  
Mild Cognitive Impairment ◽  
Longitudinal Study ◽  
Area Under The Curve ◽  
Scoring Systems ◽  
Clinical Value ◽  
Clock Drawing Test ◽  
Clock Drawing ◽  
Drawing Test ◽  
Sensitivity Specificity

ABSTRACTBackground: There is a strong demand for screening instruments for mild cognitive impairment (MCI), as a pre-stage of dementia. The clock drawing test (CDT) is widely used to screen for dementia, but the utility in screening for MCI remains uncertain. In particular, it is still questionable which scoring system is the best in order to screen for MCI. We therefore aimed to compare the utility of different CDT scoring systems for screening for MCI.Methods: In a sample of 428 subjects of the Leipzig Longitudinal Study of the Aged (LEILA 75+) study, CDT scores of different scoring systems were compared between subjects with and without MCI. Comparison of receiver operating characteristic (ROC; area under the curve, sensitivity, specificity) was performed and inter-rater reliability was calculated.Results: The CDT scores differed significantly between MCI and non-MCI subjects according to all scoring systems applied. However, ROC of the CDT scores was not adequate.Conclusions: None of the present CDT scoring systems has sufficient utility to screen reliably for MCI. The clinical value of the CDT could be improved by using semi-quantitative scoring, having a wider score range and focusing on specific details of the clock (e.g. the hands and numbers).

scholarly journals Vancomycin 24-Hour Area under the Curve/Minimum Bactericidal Concentration Ratio as a Novel Predictor of Mortality in Methicillin-Resistant Staphylococcus aureus Bacteremia

2016 ◽  
Vol 60 (5) ◽  
pp. 3070-3075 ◽  
Author(s):  
Nicholas S. Britt ◽  
Nimish Patel ◽  
Rebecca T. Horvat ◽  
Molly E. Steed
Keyword(s):  
Staphylococcus Aureus ◽  
Minimum Bactericidal Concentration ◽  
Multivariable Analysis ◽  
Area Under The Curve ◽  
Classification And Regression Tree ◽  
Mortality Data ◽  
Therapeutic Drug ◽  
Methicillin Resistant ◽  
Cart Analysis ◽  
Vancomycin Mics

ABSTRACTWhile previous studies have examined the association between vancomycin (VAN) exposure and MIC with regard to outcomes in methicillin-resistantStaphylococcus aureusbacteremia (MRSA-B), none have explored if a relationship exists with the VAN minimum bactericidal concentration (MBC). The objective of this study was to evaluate the VAN 24-h area under the curve (AUC24)/MBC ratio as a pharmacodynamic predictor of mortality. This retrospective cohort study included patients treated with VAN for MRSA-B with the primary outcome of 30-day all-cause mortality. Data collected included patient demographics, comorbidities, antimicrobial treatment data, therapeutic drug levels, and laboratory and microbiological data. Vancomycin MICs and MBCs were determined by Etest (MIC only) and broth microdilution (BMD). The vancomycin AUC24was determined by pharmacokinetic maximuma posterioriprobability Bayesian (MAP-Bayesian) analysis. The most significant breakpoint for 30-day mortality was determined by classification and regression tree (CART) analysis. The association between pharmacodynamic parameters (VAN AUC24/MICBMD, VAN AUC24/MICEtest, and AUC24/MBCBMD) and mortality were determined by χ2and multivariable Poisson regression. Overall mortality in this cohort (n= 53) was 20.8% (n= 11/53), and all corresponding MRSA blood isolates were VAN susceptible (MIC range, 0.5 to 2 μg/ml; MIC50, 1 μg/ml; MIC90, 1 μg/ml). The CART-derived breakpoints for mortality were 176 (VAN AUC24/MBC) and 334 (VAN AUC24/MICBMD). In multivariable analysis, the association between a VAN AUC24/MBC of ≥176 and survival persisted, but VAN AUC24/MICBMDvalues (≥334 or ≥400) were not associated with improved mortality. In conclusion, VAN AUC24/MBC was a more important predictor of 30-day mortality than VAN AUC24/MIC for MRSA-B.

Discriminating Malignancy in Thyroid Nodules: The Nomogram Versus the Kwak and ACR TI-RADS

2020 ◽  
Vol 163 (6) ◽  
pp. 1156-1165
Author(s):  
Juan Xiao ◽  
Qiang Xiao ◽  
Wei Cong ◽  
Ting Li ◽  
Shouluan Ding ◽  
...  
Keyword(s):  
Thyroid Nodules ◽  
Characteristic Curve ◽  
Area Under The Curve ◽  
Diagnostic Study ◽  
Diagnostic Efficiency ◽  
Training Set ◽  
Multivariable Logistic Regression Model ◽  
Predictive Values ◽  
Validation Set ◽  
Sensitivity Specificity

Objective To develop an easy-to-use nomogram for discrimination of malignant thyroid nodules and to compare diagnostic efficiency with the Kwak and American College of Radiology (ACR) Thyroid Imaging, Reporting and Data System (TI-RADS). Study Design Retrospective diagnostic study. Setting The Second Hospital of Shandong University. Subjects and Methods From March 2017 to April 2019, 792 patients with 1940 thyroid nodules were included into the training set; from May 2019 to December 2019, 174 patients with 389 nodules were included into the validation set. Multivariable logistic regression model was used to develop a nomogram for discriminating malignant nodules. To compare the diagnostic performance of the nomogram with the Kwak and ACR TI-RADS, the area under the receiver operating characteristic curve, sensitivity, specificity, and positive and negative predictive values were calculated. Results The nomogram consisted of 7 factors: composition, orientation, echogenicity, border, margin, extrathyroidal extension, and calcification. In the training set, for all nodules, the area under the curve (AUC) for the nomogram was 0.844, which was higher than the Kwak TI-RADS (0.826, P = .008) and the ACR TI-RADS (0.810, P < .001). For the 822 nodules >1 cm, the AUC of the nomogram was 0.891, which was higher than the Kwak TI-RADS (0.852, P < .001) and the ACR TI-RADS (0.853, P < .001). In the validation set, the AUC of the nomogram was also higher than the Kwak and ACR TI-RADS ( P < .05), each in the whole series and separately for nodules >1 or ≤1 cm. Conclusions When compared with the Kwak and ACR TI-RADS, the nomogram had a better performance in discriminating malignant thyroid nodules.

S-CCCapsule: Pneumonia detection in chest X-ray images using skip-connected convolutions and capsule neural network

2021 ◽  
pp. 1-25
Author(s):  
Kwabena Adu ◽  
Yongbin Yu ◽  
Jingye Cai ◽  
Victor Dela Tattrah ◽  
James Adu Ansere ◽  
...  
Keyword(s):  
Neural Network ◽  
Human Error ◽  
Medical Center ◽  
Confusion Matrix ◽  
Area Under The Curve ◽  
Dynamic Routing ◽  
Sigmoid Function ◽  
Radiological Society ◽  
Chest X Ray ◽  
Sensitivity Specificity

The squash function in capsule networks (CapsNets) dynamic routing is less capable of performing discrimination of non-informative capsules which leads to abnormal activation value distribution of capsules. In this paper, we propose vertical squash (VSquash) to improve the original squash by preventing the activation values of capsules in the primary capsule layer to shrink non-informative capsules, promote discriminative capsules and avoid high information sensitivity. Furthermore, a new neural network, (i) skip-connected convolutional capsule (S-CCCapsule), (ii) Integrated skip-connected convolutional capsules (ISCC) and (iii) Ensemble skip-connected convolutional capsules (ESCC) based on CapsNets are presented where the VSquash is applied in the dynamic routing. In order to achieve uniform distribution of coupling coefficient of probabilities between capsules, we use the Sigmoid function rather than Softmax function. Experiments on Guangzhou Women and Children’s Medical Center (GWCMC), Radiological Society of North America (RSNA) and Mendeley CXR Pneumonia datasets were performed to validate the effectiveness of our proposed methods. We found that our proposed methods produce better accuracy compared to other methods based on model evaluation metrics such as confusion matrix, sensitivity, specificity and Area under the curve (AUC). Our method for pneumonia detection performs better than practicing radiologists. It minimizes human error and reduces diagnosis time.

Sign in / Sign up

Export Citation Format

Share Document