NIMG-22. PREDICTION OF GLIOBLASTOMA CELLULAR INFILTRATION AND RECURRENCE USING MACHINE LEARNING AND MULTI-PARAMETRIC MRI ANALYSIS: RESULTS FROM THE MULTI-INSTITUTIONAL RESPOND CONSORTIUM

2021 ◽  
Vol 23 (Supplement_6) ◽  
pp. vi132-vi133
Author(s):  
Hamed Akbari ◽  
Suyash Mohan ◽  
Jose A Garcia ◽  
Anahita Fathi Kazerooni ◽  
Chiharu Sako ◽  
...  
Keyword(s):  
Odds Ratio ◽  
Tumor Recurrence ◽  
De Novo ◽  
Predictive Biomarkers ◽  
Neoplastic Cell ◽  
University Hospital ◽  
Cell Infiltration ◽  
New York University ◽  
Single Institution ◽  
Peritumoral Tissue

Abstract PURPOSE Multi-parametric MRI and artificial intelligence (AI) methods were previously used to predict peritumoral neoplastic cell infiltration and risk of future recurrence in glioblastoma, in single-institution studies. We hypothesize that important characteristics of peritumoral tissue heterogeneity captured, engineered/selected, and quantified by these methods relate to predictions generalizable in the multi-institutional ReSPOND (Radiomics Signatures for PrecisiON Diagnostics) consortium. METHODS To support further development, generalization, and clinical translation of our proposed method, we trained the AI model on a retrospective cohort of 29 de novo glioblastoma patients from the Hospital of the University of Pennsylvania (UPenn) (Male/Female:20/9, age:22-78 years) followed by evaluation on a prospective multi-institutional cohort of 84 glioblastoma patients (Male/Female:51/33, age:34-89 years) from Case Western Reserve University/University Hospitals (CWRU/UH, 25), New York University (NYU, 13), Ohio State University (OSU, 13), University Hospital Río Hortega (RH, 2), and UPenn (31). Features extracted from pre-resection MRI (T1, T1-Gd, T2, T2-FLAIR, ADC) were used to build our model predicting the spatial pattern of subsequent tumor recurrence. These predictions were evaluated against regions of pathology-confirmed post-resection recurrence. RESULTS Our model predicted the locations that later harbored tumor recurrence with sensitivity 83%, AUC 0.83 (99% CI, 0.73-0.93), and odds ratio 7.23 (99% CI, 7.09-7.37) in the prospective cohort. Odds ratio (99% CI)/AUC(99% CI) per institute were: CWRU/UH, 7.8(7.6-8.1)/0.82(0.75-0.89); NYU, 3.5(3.3-3.6)/0.84(074-0.93); OSU, 7.9(7.6-8.3)/0.8(0.67-0.94); RH, 22.7(20-25.1)/0.94(0.27-1); UPenn, 7.1(6.8-7.3)/0.83(0.76-0.91). CONCLUSION This is the first study that provides relatively extensive multi-institutional validated evidence that AI can provide good predictions of peritumoral neoplastic cell infiltration and future recurrence, by dissecting the MRI signal heterogeneity in peritumoral tissue. Our analyses leveraged the unique dataset of the ReSPOND consortium, which aims to develop and evaluate AI-based biomarkers for individualized prediction and prognostication, by moving from single-institution studies to generalizable, well-validated multi-institutional predictive biomarkers.

Sex-Related Anatomical Feasibility Differences in Endovascular Repair of Thoracoabdominal Aortic Aneurysms With Multibranched Stent-Grafts

2020 ◽  
pp. 152660282096491
Author(s):  
Alessandro Grandi ◽  
Niccolò Carta ◽  
Tommaso Cambiaghi ◽  
Victor Bilman ◽  
Germano Melissano ◽  
...  
Keyword(s):  
Vascular Access ◽  
Endovascular Repair ◽  
Odds Ratio ◽  
Multivariable Analysis ◽  
Aortic Aneurysms ◽  
Clinical Factors ◽  
Stent Grafts ◽  
Single Institution ◽  
Confounding Variables ◽  
Instructions For Use

Purpose: To evaluate the potential anatomical feasibility of using the off-the-shelf multibranched Zenith t-Branch for the treatment of thoracoabdominal aortic aneurysms (TAAAs) in female patients. Materials and Methods: A total of 268 patients (median age 68 years; 69 women) with degenerative TAAA treated at a single institution by means of open or endovascular repair between 2007 and 2019 were retrospectively analyzed to determine the feasibility of using the Zenith t-Branch based on the manufacturer’s instructions for use. The factors determining overall anatomical feasibility were divided into vascular access, aortic anatomy, and visceral vessels. The results were stratified by sex and compared. A logistic regression model was constructed to determine any association between feasibility and clinical factors or potential confounding variables; results are expressed as the odds ratio (OR) with 95% confidence interval (CI). Results: The overall anatomical feasibility was 39% (22% women vs 45% men, p=0.001). The feasibility was negatively influenced by female sex (p<0.001) in multivariable analysis (OR 2.9, 95% CI 1.5 to 5.4, p=0.001). Vascular access feasibility was 82% (61% women vs 89% men, p<0.001). Aorta feasibility was 65% (52% women vs 69% men, p<0.001), and visceral vessel feasibility was 74% (78% women vs 73% men, p=0.260). An access diameter ≤8.5 mm excluded 17% of the patients (39% women vs 9% men, p<0.001). The aortic feasibility was limited by the infrarenal aortic diameter in 16% of patients (45% women vs 6% men, p<0.001) and the aortic lumen at the visceral vessels in 17% patients (19% women vs 17% men, p=0.741). The visceral vessel feasibility was mainly limited by inadequate numbers or diameters of target vessels. Location and orientation of the target vessels were adequate in 96% of patients. Conclusion: A little more than a third of an all-comers cohort of patients with degenerative TAAA could have been treated with on-label use of the Zenith t-Branch. However, only 22% of women could have been treated because of sex-related anatomical limitations. New generations of multibranched devices should address these differences.

Using the diagnostic odds ratio to select multivariate sequential patterns in order to build an interpretable pattern-based classifier in a clinical domain (Preprint)

2021 ◽  
Author(s):  
Isidoro J. Casanova ◽  
Manuel Campos ◽  
Jose M. Juarez ◽  
Antonio Gomariz ◽  
Marta Lorente-Ros ◽  
...  
Keyword(s):  
Confidence Intervals ◽  
Odds Ratio ◽  
Diagnostic Odds Ratio ◽  
University Hospital ◽  
Classification Model ◽  
Sequential Patterns ◽  
Pattern Selection ◽  
Essential Property ◽  
Frequency Properties ◽  
Clinical Domain

BACKGROUND It is important to exploit all available data on patients in settings such as Intensive Care Burn Units (ICBUs), where several variables are recorded over time. It is possible to take advantage of the multivariate patterns that model the evolution of patients in order to predict their survival. However, pattern discovery algorithms generate a large number of patterns, of which only some are relevant for classification. The interpretability of the model is, moreover, an essential property in the clinical domain. OBJECTIVE We propose to use the Diagnostic Odds Ratio (DOR) to select the multivariate sequential patterns used in the classification in a clinical domain, rather than employing frequency properties. This makes it possible to employ a terminology closer to the language of clinicians, in which a pattern is considered to be a risk factor or to have a protection factor. METHODS We employ data obtained from the ICBU at the University Hospital of Getafe, where six temporal variables for 465 patients were registered every day during 5 days, and to model the evolution of these clinical variables we use multivariate sequential patterns. We compare four ways in which to employ the DOR for pattern selection: 1) We use it as a threshold in order to select patterns with a minimum DOR; 2) We select patterns whose differential DORs are higher than a threshold as regards their extensions; 3) We select patterns whose DOR confidence intervals do not overlap; and 4) We propose the combination of threshold and non-overlapping confidence intervals in order to select the most discriminative patterns. As a baseline, we compare our proposals with Jumping Emerging Patterns (JEPs), one of the most frequently used techniques for pattern selection that utilize frequency properties. RESULTS We have compared the number and length of the patterns eventually selected, classification performance, and pattern and model interpretability. We show that discretization has a great impact on the accuracy of the classification model, but that a trade off must be found between classification accuracy and the physicians' capacity to interpret the patterns obtained. We have, therefore, opted to use expert discretization without losing too much accuracy. We have also identified that the experiments combining threshold and non-overlapping confidence intervals (Option 4) obtained the fewest number of patterns but also with the smallest size, thus implying the loss of an acceptable accuracy as regards clinician interpretation. CONCLUSIONS A method for the classification of patients’ survival can benefit from the use of sequential patterns, since these patterns consider knowledge about the temporal evolution of the variables in the case of ICBU. We have proved that the DOR can be used in several ways, and that it is a suitable measure with which to select discriminative and interpretable quality patterns.

Mendelian Randomization Focused Analysis of Vitamin D on the Secondary Prevention of Ischemic Stroke

Stroke ◽  
2021 ◽  
Author(s):  
Yap-Hang Chan ◽  
C. Mary Schooling ◽  
Jie Zhao ◽  
Shiu-Lun Au Yeung ◽  
Jo Jo Hai ◽  
...  
Keyword(s):  
Myocardial Infarction ◽  
Vitamin D ◽  
Ischemic Stroke ◽  
Risk Score ◽  
Genetic Risk ◽  
Odds Ratio ◽  
Mendelian Randomization ◽  
De Novo ◽  
Genetic Risk Score ◽  
Ischemic Disease

Background and Purpose: Experimental studies showed vitamin D (Vit-D) could promote vascular regeneration and repair. Prior randomized studies had focused mainly on primary prevention. Whether Vit-D protects against ischemic stroke and myocardial infarction recurrence among subjects with prior ischemic insults was unknown. Here, we dissected through Mendelian randomization any effect of Vit-D on the secondary prevention of recurrent ischemic stroke and myocardial infarction. Methods: Based on a genetic risk score for Vit-D constructed from a derivation cohort sample (n=5331, 45% Vit-D deficient, 89% genotyped) via high-throughput exome-chip screening of 12 prior genome-wide association study–identified genetic variants of Vit-D mechanistic pathways ( rs2060793 , rs4588 , and rs7041 ; F statistic, 73; P <0.001), we performed a focused analysis on prospective recurrence of myocardial infarction (MI) and ischemic stroke in an independent subsample with established ischemic disease (n=441, all with prior first ischemic event; follow-up duration, 41.6±14.3 years) under a 2-sample, individual-data, prospective Mendelian randomization approach. Results: In the ischemic disease subsample, 11.1% (n=49/441) had developed recurrent ischemic stroke or MI and 13.3% (n=58/441) had developed recurrent or de novo ischemic stroke/MI. Kaplan-Meier analyses showed that genetic risk score predicted improved event-free survival from recurrent ischemic stroke or MI (log-rank, 13.0; P =0.001). Cox regression revealed that genetic risk score independently predicted reduced risk of recurrent ischemic stroke or MI combined (hazards ratio, 0.62 [95% CI, 0.48–0.81]; P <0.001), after adjusted for potential confounders. Mendelian randomization supported that Vit-D is causally protective against the primary end points of recurrent ischemic stroke or MI (Wald estimate: odds ratio, 0.55 [95% CI, 0.35–0.81]) and any recurrent or de novo ischemic stroke/MI (odds ratio, 0.64 [95% CI, 0.42–0.91]) and recurrent MI alone (odds ratio, 0.52 [95% CI, 0.30–0.81]). Conclusions: Genetically predicted lowering in Vit-D level is causal for the recurrence of ischemic vascular events in persons with prior ischemic stroke or MI.

scholarly journals PORTAL GASTROPATHY

2018 ◽  
Vol 25 (12) ◽  
pp. 1899-1904
Author(s):  
Aftab Abbasi ◽  
Sana Naz ◽  
Faisal Razi
Keyword(s):  
Liver Cirrhosis ◽  
Inflammatory Cell ◽  
Inflammatory Cell Infiltration ◽  
University Hospital ◽  
Cell Infiltration ◽  
Spearman Correlation ◽  
Gastric Glands ◽  
Positive Correlation ◽  
Pugh Class ◽  
Tissue Specimens

Objectives: To study the characteristic histological features of Portal gastropathy and its correlation with Child-Pugh Score in liver cirrhosis patients. Study Design: Cross sectional study. Place and Duration: Department of Anatomy and Gastroenterology Unit, Isra University Hospital from June to December 2012. Subjects and Methods: Gastric biopsies from 85 cases of liver cirrhosis with portal gastropathy were collected by non probabilityconvenient sampling. Olympus XQ 140 (version 3) was used for 2 mm thick tissue specimens by punch biopsy. Tissue pieces were preserved in 10% formalin. 3-5 μ thick tissue specimens were stained with H & E for microscopic examination. Results: Portal Gastropathy was noted in 91.7% of total study subjects. Gastric glands showed increase counts and increased size noted in 60.12% and 57.65% of cases respectively. Pyloric antrum revealed inflammatory cell infiltration of lamina propria. 95.2% inflammatory cells comprised of lymphocytes. Capillary congestion and edema was noted in 9.4% of cases. Spearman correlation showed positive correlation of Portal gastropathy and Child Pugh Class (CPC) score (R2=0.5244, p=0.0001).Conclusion: Portal Gastropathy was noted in 91.7%. Histological showed increase in size, length and count of gastric glands, capillary congestion and inflammatory cell infiltration. Spearman correlation showed positive correlation of Portal gastropathy and Child Pugh Class (CPC) score (R2=0.5244, p=0.0001). 

scholarly journals The prenatal diagnosis and clinical outcomes of fetuses with 15q11.2 copy number variants: a case series of 36 patients

2021 ◽  
Author(s):  
Jessica Kang ◽  
Chien Nan Lee ◽  
Yi-Ning Su ◽  
Ming-Wei Lin ◽  
Yi-Yun Tai ◽  
...  
Keyword(s):  
Pregnant Women ◽  
Developmental Delay ◽  
Copy Number ◽  
De Novo ◽  
Copy Number Variants ◽  
Case Series ◽  
Copy Number Variant ◽  
University Hospital ◽  
Limited Information ◽  
Ultrasound Findings

Objective: The prenatal genetic counseling of fetus diagnosed with the 15q11.2 copy number variant (CNV) involving the BP1-BP2 region has been difficult due to limited information and controversial opinion on prognosis. Design: Case series. Setting: This study uses data from National Taiwan University Hospital. Sample: Data of 36 pregnant women who underwent prenatal microarray analysis from 2012 to 2017 and were assessed at National Taiwan University Hospital. Methods: Data were collected by reviewing patients’ medical record. Comparison of patient characteristics, prenatal ultrasound findings and postnatal outcomes between different cases involving the 15q11.2 BP1-BP2 region were presented. Main outcome measured: Postnatal prognosis. Results: Out of the 36 patients diagnosed with CNVs involving the BP1-BP2 region, 5 were diagnosed with microduplication and 31 with microdeletion. Abnormal ultrasound findings were recorded in 12 cases prenatally. De novo microduplications were observed in 25% of the cases and microdeletions were found in 14%. Amongst the cases, 10 pregnant women received termination of pregnancy and 26 gave birth to healthy individuals (27 babies in total). Conclusion: The prognoses of 15q11.2 CNVs were controversial and recent studies have revealed its connection with developmental delay and autism. In our study, no obvious developmental delay or neurological disorders were detected postnatally in the 1 case of 15q11.2 microduplication and 25 cases of microdeletion.

Venous Thromboembolism in Patients With Thrombocytopenia: Risk Factors, Treatment, and Outcome

2015 ◽  
Vol 23 (4) ◽  
pp. 345-350 ◽  
Author(s):  
Jens Kristian Baelum ◽  
Espen Ellingsen Moe ◽  
Mads Nybo ◽  
Pernille Just Vinholt
Keyword(s):  
Risk Factors ◽  
Platelet Count ◽  
Venous Thromboembolism ◽  
Blood Cells ◽  
Medical Records ◽  
Odds Ratio ◽  
Conditional Logistic Regression ◽  
University Hospital ◽  
Hospital Department ◽  
Bleeding Episodes

Background: Venous thromboembolism (VTE) is a frequent and potentially lethal condition. Venous thrombi are mainly constituted of fibrin and red blood cells, but platelets also play an important role in VTE formation. Information about VTE in patients with thrombocytopenia is, however, missing. Objectives: To identify VTE risk factors and describe treatment and outcome (bleeding episodes and mortality) in patients with thrombocytopenia. Patients/Methods: Patients with thrombocytopenia (platelet count <100 × 109/L) admitted to Odense University Hospital, Denmark, between April 2000 and April 2012 were included. Fifty cases had experienced VTE. Controls without VTE were matched 3:1 with cases on sex and hospital department. Medical records were examined, and data were analyzed using conditional logistic regression. Results: In multivariate analysis, platelet count <50 × 109/L (odds ratio [OR] 0.22, P < .05) and chronic liver disease (OR 0.05, 95% confidence interval [CI] 0.01-0.58) reduced the risk of VTE. Surgery (OR 6.44, 95% CI 1.37-30.20) and previous thromboembolism (OR 6.16, 95% CI 1.21-31.41) were associated with an increased VTE risk. Ninety-two percent of cases were treated with anticoagulants. There was no difference in bleeding incidence between cases and controls. Conclusions: Several known VTE risk factors also seems to apply in patients with thrombocytopenia. Also, patients with thrombocytopenia may be VTE risk stratified based on platelet count and comorbidities. Finally, patients having thrombocytopenia with VTE seem to be safely treated with anticoagulants without increased occurrence of bleeding.

Sign in / Sign up

Export Citation Format

Share Document