Categories
Nevin Manimala Statistics

The effect of CT texture-based analysis using machine learning approaches on radiologists’ performance in differentiating focal-type autoimmune pancreatitis and pancreatic duct carcinoma

Jpn J Radiol. 2022 Jun 21. doi: 10.1007/s11604-022-01298-7. Online ahead of print.

ABSTRACT

PURPOSE: To develop a support vector machine (SVM) classifier using CT texture-based analysis in differentiating focal-type autoimmune pancreatitis (AIP) and pancreatic duct carcinoma (PD), and to assess the radiologists’ diagnostic performance with or without SVM.

MATERIALS AND METHODS: This retrospective study included 50 patients (20 patients with focal-type AIP and 30 patients with PD) who underwent dynamic contrast-enhanced CT. Sixty-two CT texture-based features were extracted from 2D images of the arterial and portal phase CTs. We conducted data compression and feature selections using principal component analysis (PCA) and produced the SVM classifier. Four readers participated in this observer performance study and the statistical significance of differences with and without the SVM was assessed by receiver operating characteristic (ROC) analysis.

RESULTS: The SVM performance indicated a high performance in differentiating focal-type AIP and PD (AUC = 0.920). The AUC for all 4 readers increased significantly from 0.827 to 0.911 when using the SVM outputs (p = 0.010). The AUC for inexperienced readers increased significantly from 0.781 to 0.905 when using the SVM outputs (p = 0.310). The AUC for experienced readers increased from 0.875 to 0.912 when using the SVM outputs, however, there was no significant difference (p = 0.018).

CONCLUSION: The use of SVM classifier using CT texture-based features improved the diagnostic performance for differentiating focal-type AIP and PD on CT.

PMID:35727458 | DOI:10.1007/s11604-022-01298-7

Categories
Nevin Manimala Statistics

Clinical Analysis of Primary Tracheobronchial Tumors in Children and Evaluation of the Predicting Models for Mucoepidermoid Carcinoma

Curr Med Sci. 2022 Jun 21. doi: 10.1007/s11596-022-2595-3. Online ahead of print.

ABSTRACT

OBJECTIVE: To determine the clinical characteristics and prognosis of primary tracheobronchial tumors (PTTs) in children, and to explore the most common tumor identification methods.

METHODS: The medical records of children with PTTs who were hospitalized at the Children’s Hospital of Chongqing Medical University from January 1995 to January 2020 were reviewed retrospectively. The clinical features, imaging, treatments, and outcomes of these patients were statistically analyzed. Machine learning techniques such as Gaussian naïve Bayes, support vector machine (SVM) and decision tree models were used to identify mucoepidermoid carcinoma (ME).

RESULTS: A total of 16 children were hospitalized with PTTs during the study period. This included 5 (31.3%) children with ME, 3 (18.8%) children with inflammatory myofibroblastic tumors (IMT), 2 children (12.5%) with sarcomas, 2 (12.5%) children with papillomatosis and 1 child (6.3%) each with carcinoid carcinoma, adenoid cystic carcinoma (ACC), hemangioma, and schwannoma, respectively. ME was the most common tumor type and amongst the 3 ME recognition methods, the SVM model showed the best performance. The main clinical symptoms of PPTs were cough (81.3%), breathlessness (50%), wheezing (43.8%), progressive dyspnea (37.5%), hemoptysis (37.5%), and fever (25%). Of the 16 patients, 7 were treated with surgery, 8 underwent bronchoscopic tumor resection, and 1 child died. Of the 11 other children, 3 experienced recurrence, and the last 8 remained disease-free. No deaths were observed during the follow-up period.

CONCLUSION: PTT are very rare in children and the highest percentage of cases is due to ME. The SVM model was highly accurate in identifying ME. Chest CT and bronchoscopy can effectively diagnose PTTs. Surgery and bronchoscopic intervention can both achieve good clinical results and the prognosis of the 11 children that were followed up was good.

PMID:35727419 | DOI:10.1007/s11596-022-2595-3

Categories
Nevin Manimala Statistics

Negative interaction between nitrates and remote ischemic preconditioning in patients undergoing cardiac surgery: the ERIC-GTN and ERICCA studies

Basic Res Cardiol. 2022 Jun 21;117(1):31. doi: 10.1007/s00395-022-00938-3.

ABSTRACT

Remote ischaemic preconditioning (RIPC) using transient limb ischaemia failed to improve clinical outcomes following cardiac surgery and the reasons for this remain unclear. In the ERIC-GTN study, we evaluated whether concomitant nitrate therapy abrogated RIPC cardioprotection. We also undertook a post-hoc analysis of the ERICCA study, to investigate a potential negative interaction between RIPC and nitrates on clinical outcomes following cardiac surgery. In ERIC-GTN, 185 patients undergoing cardiac surgery were randomized to: (1) Control (no RIPC or nitrates); (2) RIPC alone; (3); Nitrates alone; and (4) RIPC + Nitrates. An intravenous infusion of nitrates (glyceryl trinitrate 1 mg/mL solution) was commenced on arrival at the operating theatre at a rate of 2-5 mL/h to maintain a mean arterial pressure between 60 and 70 mmHg and was stopped when the patient was taken off cardiopulmonary bypass. The primary endpoint was peri-operative myocardial injury (PMI) quantified by a 48-h area-under-the-curve high-sensitivity Troponin-T (48 h-AUC-hs-cTnT). In ERICCA, we analysed data for 1502 patients undergoing cardiac surgery to investigate for a potential negative interaction between RIPC and nitrates on clinical outcomes at 12-months. In ERIC-GTN, RIPC alone reduced 48 h-AUC-hs-cTnT by 37.1%, when compared to control (ratio of AUC 0.629 [95% CI 0.413-0.957], p = 0.031), and this cardioprotective effect was abrogated in the presence of nitrates. Treatment with nitrates alone did not reduce 48 h-AUC-hs-cTnT, when compared to control. In ERICCA there was a negative interaction between nitrate use and RIPC for all-cause and cardiovascular mortality at 12-months, and for risk of peri-operative myocardial infarction. RIPC alone reduced the risk of peri-operative myocardial infarction, compared to control, but no significant effect of RIPC was demonstrated for the other outcomes. When RIPC and nitrates were used together they had an adverse impact in patients undergoing cardiac surgery with the presence of nitrates abrogating RIPC-induced cardioprotection and increasing the risk of mortality at 12-months post-cardiac surgery in patients receiving RIPC.

PMID:35727392 | DOI:10.1007/s00395-022-00938-3

Categories
Nevin Manimala Statistics

Incidence of acute myocardial injury and its association with left and right ventricular systolic dysfunction in critically ill COVID-19 patients

Ann Intensive Care. 2022 Jun 21;12(1):56. doi: 10.1186/s13613-022-01030-8.

ABSTRACT

BACKGROUND: Previous studies have found an increase in cardiac troponins (cTns) and echocardiographic abnormalities in patients with COVID-19 and reported their association with poor clinical outcomes. Whether acute injury occurs during the course of critical care and if it is associated with cardiac function is unknown. The purpose of this study was to document the incidence of acute myocardial injury (AMInj) and echocardiographically defined left ventricular (LV) and right ventricular (RV) systolic dysfunction in consecutive patients admitted to an intensive care unit (ICU) for COVID-19. The relationship between AMInj and echocardiographic abnormalities during the first 14 days of ICU admission was studied. Finally, the association between echocardiographic findings, AMInj and clinical outcome was evaluated.

METHODS: Seventy-four consecutive patients (≥18 years) admitted to the ICU at Linköping University Hospital between 19 Mar 2020 and 31 Dec 2020 for COVID-19 were included. High-sensitivity troponin-T (hsTnT) was measured daily for up to 14 days. Transthoracic echocardiography was conducted within 72 h of ICU admission. Acute myocardial injury was defined as an increased hsTnT > 14ng/l and a > 20% absolute change with or without ischaemic symptoms. LV and RV systolic dysfunction was defined as at least 2 abnormal indicators of systolic function specified by consensus guidelines.

RESULTS: Increased hsTnT was observed in 59% of patients at ICU admission, and 82% developed AMInj with peak levels at 8 (3-13) days after ICU admission. AMInj was not statistically significantly associated with 30-day mortality but was associated with an increased duration of invasive mechanical ventilation (10 (3-13) vs. 5 days (0-9), p=0.001) as well as ICU length of stay (LOS) (19.5 (11-28) vs. 7 days (5-13), p=0.015). After adjustment for SAPS-3 and admission SOFA score, the effect of AMInj was significant only for the duration of mechanical ventilation (p=0.030). The incidence of LV and RV dysfunction was 28% and 22%, respectively. Only indices of LV and RV longitudinal contractility (mitral and tricuspid annular plane systolic excursion) were associated with AMInj. Echocardiographic parameters were not associated with clinical outcome.

CONCLUSIONS: Myocardial injury is common in critically ill patients with COVID-19, with AMInj developing in more than 80% after ICU admission. In contrast, LV and RV dysfunction occurred in approximately one-quarter of patients. AMInj was associated with an increased need for mechanical ventilation and ICU LOS but neither AMInj nor ventricular dysfunction was significantly associated with mortality.

PMID:35727386 | DOI:10.1186/s13613-022-01030-8

Categories
Nevin Manimala Statistics

miRNAs as therapeutic predictors and prognostic biomarkers of neoadjuvant chemotherapy in breast cancer: a systematic review and meta-analysis

Breast Cancer Res Treat. 2022 Jun 21. doi: 10.1007/s10549-022-06642-z. Online ahead of print.

ABSTRACT

PURPOSE: Accumulating evidence has shown that microRNAs (miRNAs) are promising biomarkers of neoadjuvant chemotherapy (NAC) response in breast cancer (BC). However, their predictive roles remain controversial. Thus, this systematic review and meta-analysis aimed to describe the role of miRNA expression in NAC response and prognosis in BC to increase statistical power and improve translation.

METHODS: A systematic review of electronic databases for relevant studies was conducted following PRISMA guidelines. Data were extracted, collated, and combined by odds ratio (OR) and hazard ratio (HR) with 95% confidence intervals (CIs) to estimate the strength of the associations.

RESULTS: Of the 560 articles screened, 59 studies were included in our systematic review, and 5 studies were included in the subsequent meta-analysis. Sixty of 123 miRNAs were found to be related with NAC response and an elevated baseline miR-7 level in tissues was associated with a higher pathological complete response rate (OR 5.63; 95% CI 2.15-14.79; P = 0.0004). The prognostic value of 39 miRNAs was also studied. Of them, 26 miRNAs were found to be associated with survival. Pooled HRs indicated that patients with increased levels of serum miR-21 from baseline to the end of the second NAC cycle and from baseline to the end of NAC had a worse disease-free survival than those with decreased levels.

CONCLUSION: Our results highlight that a large number of miRNAs have possible associations with NAC response and prognosis in BC patients. Further well-designed studies are needed to elucidate the molecular mechanisms underlying these associations.

PMID:35727379 | DOI:10.1007/s10549-022-06642-z

Categories
Nevin Manimala Statistics

Safety of sequential immune checkpoint inhibitors after prior immune therapy

J Cancer Res Clin Oncol. 2022 Jun 21. doi: 10.1007/s00432-022-04137-4. Online ahead of print.

ABSTRACT

BACKGROUND: The use of immune checkpoint inhibitors (ICI) has transformed cancer treatment. Subsequent ICI use has become increasingly common following disease progression. We aim to evaluate the safety and tolerability of the sequential ICI treatment modality.

METHODS: Retrospective review of confirmed carcinoma from January 2014 to December 2018. Patients were categorized into “initial ICI arm” and “sequential ICI arm” defined as patients receiving single, dual or chemo-immunotherapy ICI following an initial ICI regimen. Primary outcome was the development of a new or recurrent immune related adverse event (irAE) during sequential therapy. Secondary outcomes were the number of cycles prior to the development of irAE and grade of irAE.

RESULTS: A total of 483 patients received ICI during the timeframe. Of those, 22 patients received sequential ICI. The diagnoses included ten lung cancer, seven melanoma, four renal cell carcinoma and one bladder cancer. 16 patients received single agent ICI following the initial ICI, three patients received dual ICI following the initial ICI, one patient received chemotherapy-immunotherapy following initial ICI, and two patients received chemo-immunotherapy after dual ICI. Four patients developed new irAE and one patient developed the same irAE on sequential treatment. A higher proportion of patients experienced grade 3 irAE in the sequential arm compared to the initial ICI arm (p = 0.03). No statistical difference was found between the development of irAE and the number of cycles prior to development of irAE in either treatment groups (p = 0.5).

CONCLUSION: Our data shows overall safety of sequencing ICI when close monitoring was employed.

PMID:35727369 | DOI:10.1007/s00432-022-04137-4

Categories
Nevin Manimala Statistics

The unidirectional prosaccade switch-cost: no evidence for the passive dissipation of an oculomotor task-set inertia

Exp Brain Res. 2022 Jun 21. doi: 10.1007/s00221-022-06394-8. Online ahead of print.

ABSTRACT

Cognitive flexibility is a core component of executive function and supports the ability to ‘switch’ between different tasks. Our group has examined the cost associated with switching between a prosaccade (i.e., a standard task requiring a saccade to veridical target location) and an antisaccade (i.e., a non-standard task requiring a saccade mirror-symmetrical to veridical target) in predictable (i.e., AABB) and unpredictable (e.g., AABAB…) switching paradigms. Results have shown that reaction times (RTs) for a prosaccade preceded by an antisaccade (i.e., task-switch trial) are longer than when preceded by its same task-type (i.e., task-repeat trial), whereas RTs for antisaccade task-switch and task-repeat trials do not differ. The asymmetrical switch-cost has been attributed to an antisaccade task-set inertia that proactively delays a subsequent prosaccade (i.e., the unidirectional prosaccade switch-cost). A salient question arising from previous work is whether the antisaccade task-set inertia passively dissipates or persistently influences prosaccade RTs. Accordingly, participants completed separate AABB (i.e., A = prosaccade, B = antisaccade) task-switching conditions wherein the preparation interval for each trial was ‘short’ (1000-2000 ms; i.e., the timeframe used in previous work), ‘medium’ (3000-4000 ms) and ‘long’ (5000-6000 ms). Results demonstrated a reliable prosaccade switch-cost for each condition (ps < 0.02) and two one-sided test statistics indicated that switch cost magnitudes were within an equivalence boundary (ps < 0.05). Hence, null and equivalence tests demonstrate that an antisaccade task-set inertia does not passively dissipate and represents a temporally persistent feature of oculomotor control.

PMID:35727365 | DOI:10.1007/s00221-022-06394-8

Categories
Nevin Manimala Statistics

Autoencoders reloaded

Biol Cybern. 2022 Jun 21. doi: 10.1007/s00422-022-00937-6. Online ahead of print.

ABSTRACT

In Bourlard and Kamp (Biol Cybern 59(4):291-294, 1998), it was theoretically proven that autoencoders (AE) with single hidden layer (previously called “auto-associative multilayer perceptrons”) were, in the best case, implementing singular value decomposition (SVD) Golub and Reinsch (Linear algebra, Singular value decomposition and least squares solutions, pp 134-151. Springer, 1971), equivalent to principal component analysis (PCA) Hotelling (Educ Psychol 24(6/7):417-441, 1993); Jolliffe (Principal component analysis, springer series in statistics, 2nd edn. Springer, New York ). That is, AE are able to derive the eigenvalues that represent the amount of variance covered by each component even with the presence of the nonlinear function (sigmoid-like, or any other nonlinear functions) present on their hidden units. Today, with the renewed interest in “deep neural networks” (DNN), multiple types of (deep) AE are being investigated as an alternative to manifold learning Cayton (Univ California San Diego Tech Rep 12(1-17):1, 2005) for conducting nonlinear feature extraction or fusion, each with its own specific (expected) properties. Many of those AE are currently being developed as powerful, nonlinear encoder-decoder models, or used to generate reduced and discriminant feature sets that are more amenable to different modeling and classification tasks. In this paper, we start by recalling and further clarifying the main conclusions of Bourlard and Kamp (Biol Cybern 59(4):291-294, 1998), supporting them by extensive empirical evidences, which were not possible to be provided previously (in 1988), due to the dataset and processing limitations. Upon full understanding of the underlying mechanisms, we show that it remains hard (although feasible) to go beyond the state-of-the-art PCA/SVD techniques for auto-association. Finally, we present a brief overview on different autoencoder models that are mainly in use today and discuss their rationale, relations and application areas.

PMID:35727351 | DOI:10.1007/s00422-022-00937-6

Categories
Nevin Manimala Statistics

Time-evolving psychological processes over repeated decisions

Psychol Rev. 2022 Apr;129(3):438-456. doi: 10.1037/rev0000351.

ABSTRACT

Many psychological experiments have subjects repeat a task to gain the statistical precision required to test quantitative theories of psychological performance. In such experiments, time-on-task can have sizable effects on performance, changing the psychological processes under investigation. Most research has either ignored these changes, treating the underlying process as static, or sacrificed some psychological content of the models for statistical simplicity. We use particle Markov chain Monte-Carlo methods to study psychologically plausible time-varying changes in model parameters. Using data from three highly cited experiments, we find strong evidence in favor of a hidden Markov switching process as an explanation of time-varying effects. This embodies the psychological assumption of “regime switching,” with subjects alternating between different cognitive states representing different modes of decision-making. The switching model explains key long- and short-term dynamic effects in the data. The central idea of our approach can be applied quite generally to quantitative psychological theories, beyond the models and datasets that we investigate. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

PMID:35727307 | DOI:10.1037/rev0000351

Categories
Nevin Manimala Statistics

The role of psychological capital in the relationship between nurses’ job satisfaction and turnover intention

Perspect Psychiatr Care. 2022 Jun 21. doi: 10.1111/ppc.13128. Online ahead of print.

ABSTRACT

PURPOSE: To find out whether psychological capital has a mediating role in the relationship between job satisfaction and turnover intention on nurses.

DESIGN AND METHODS: This study is a cross-sectional descriptive study. 466 nurses participated in this study. Data were collected via paper-and-pencil format using Minnesota Satisfaction Questionnaire, Psychological Capital Questionnaire and Turnover Intention Scale.

CONCLUSIONS: Our findings revealed effects of poor psychological capital among nurses. The effect of psychological capital was statistically significant and it partially mediated job satisfaction and turnover intention.

PRACTICE IMPLICATIONS: Initiative sand training programs should be planned to improve nurses’ attitudes towards psychological capital levels.

PMID:35726709 | DOI:10.1111/ppc.13128