Machine Learning in Differentiating Gliomas from Primary CNS Lymphomas: A Systematic Review, Reporting Quality, and Risk of Bias Assessment

BACKGROUND

Differentiating gliomas and primary CNS lymphoma represents a diagnostic challenge with important therapeutic ramifications. Biopsy is the preferred method of diagnosis, while MR imaging in conjunction with machine learning has shown promising results in differentiating these tumors.

PURPOSE

Our aim was to evaluate the quality of reporting and risk of bias, assess data bases with which the machine learning classification algorithms were developed, the algorithms themselves, and their performance.

DATA SOURCES

Ovid EMBASE, Ovid MEDLINE, Cochrane Central Register of Controlled Trials, and the Web of Science Core Collection were searched according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.

STUDY SELECTION

From 11,727 studies, 23 peer-reviewed studies used machine learning to differentiate primary CNS lymphoma from gliomas in 2276 patients.

DATA ANALYSIS

Characteristics of data sets and machine learning algorithms were extracted. A meta-analysis on a subset of studies was performed. Reporting quality and risk of bias were assessed using the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) and Prediction Model Study Risk Of Bias Assessment Tool.

DATA SYNTHESIS

The highest area under the receiver operating characteristic curve (0.961) and accuracy (91.2%) in external validation were achieved by logistic regression and support vector machines models using conventional radiomic features. Meta-analysis of machine learning classifiers using these features yielded a mean area under the receiver operating characteristic curve of 0.944 (95% CI, 0.898-0.99). The median TRIPOD score was 51.7%. The risk of bias was high for 16 studies.

LIMITATIONS

Exclusion of abstracts decreased the sensitivity in evaluating all published studies. Meta-analysis had high heterogeneity.

CONCLUSIONS

Machine learning-based methods of differentiating primary CNS lymphoma from gliomas have shown great potential, but most studies lack large, balanced data sets and external validation. Assessment of the studies identified multiple deficiencies in reporting quality and risk of bias. These factors reduce the generalizability and reproducibility of the findings.

[Contrastive analysis about the postoperative clinical characteristics of elderly patients with colorectal cancer in different age groups]

Objective: To explore the postoperative clinical characteristics of elderly patients with colorectal cancer at different ages. Methods: Retrospective analysis was performed on the clinical data of 720 elderly patients with Colorectal Cancer in Beijing Hospitals from January 2012 to December 2019. There were 411 males and 309 females with a median age of 74 years. We divided the patients into young-old, old-old, oldest-old colorectal cancer patient groups and used chi-square comparative analysis of different groups of patients with clinical disease characteristics. Results: The oldest-old colorectal cancer patients tended to have normal body mass index (BMI), and the site of the disease shifted to the right. The incidence of concomitant diseases such as heart disease and hypertension increases gradually with age, and the incidence of diabetes is highest in old-old colorectal cancer patients. The proportion of open surgery was higher in the oldest-old group, but the operation time was shorter than the other two groups. In addition, the incidence of postoperative complications in elderly patients with colorectal cancer gradually increases with age, especially cardiac complications and other complications such as pneumonia, deep venous thrombosis of lower limbs, urinary retention, urinary tract infection, renal failure, cerebral hemorrhage, cerebral infarction, and so on. Conclusions: The BMI of the oldest-old patients tended to be normal, and the site of the disease shifted to the right. The incidence of heart disease, hypertension, and other concomitant diseases and postoperative complications in elderly patients with colorectal cancer gradually increase with age. Thus, the choice of treatment should be more individualized for elderly patients with colorectal cancer, and more attention should be paid to perioperative management.

Low Serum Dehydroepiandrosterone and Dehydroepiandrosterone Sulfate Are Associated With Coronary Heart Disease in Men With Type 2 Diabetes Mellitus

AIMS

Sex hormones play an important role in the pathogenesis of cardiovascular disease (CVD). This cross-sectional study aimed to explore the associations of dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS) with coronary heart disease (CHD) and stroke in middle-aged and elderly patients with type 2 diabetes mellitus (T2DM).

MATERIALS AND METHODS

A total of 995 patients with T2DM were included in the study analysis. Serum levels of DHEA and DHEAS were quantified using liquid chromatography-tandem mass spectrometry. Binary logistic regression analyses were performed to assess the associations of DHEA and DHEAS with CHD and stroke. Receiver operating characteristic (ROC) curve analysis was performed to determine the optimal DHEA and DHEAS cutoff values for the detection of CHD in men with T2DM.

RESULTS

In men with T2DM, after adjustment for potential confounders in model 3, the risk of CHD decreased with an increasing serum DHEA level [odds ratio (OR) = 0.38, quartile 4 vs. quartile 1; 95% confidence interval (CI) = 0.16-0.90; p = 0.037 for trend). Consistently, when considered as a continuous variable, this association remained significant in the fully adjusted model (OR = 0.59, 95% CI = 0.40-0.87, p < 0.05). When taken as a continuous variable in model 3, serum DHEAS level was also inversely related to the risk of CHD among men (OR = 0.56, 95% CI = 0.38-0.82, p < 0.05). Similarly, this relationship remained statistically significant when DHEAS was categorized into quartiles (OR = 0.27, quartile 4 vs. quartile 1; 95% CI = 0.11-0.67; p = 0.018 for trend). ROC curve analyses revealed that the optimal cutoff values to detect CHD in men with T2DM were 6.43 nmol/L for DHEA and 3.54 μmol/L for DHEAS. In contrast, no significant associations were found between DHEA and DHEAS on the one hand and stroke on the other in men and women with T2DM (all p > 0.05).

CONCLUSIONS

Serum DHEA and DHEAS were significantly and negatively associated with CHD in middle-aged and elderly men with T2DM. This study suggests potential roles of DHEA and DHEAS in CHD pathogenesis.

Low Serum Total Testosterone Is Associated with Non-Alcoholic Fatty Liver Disease in Men but Not in Women with Type 2 Diabetes Mellitus

MATERIALS AND METHODS

There were 1155 patients with T2DM included in the analysis. Serum levels of total testosterone and the precursors of androgens, including androstenedione, DHEA, and DHEAS, were quantified using liquid chromatography-tandem mass spectrometry assays.

RESULTS

The risk of NAFLD decreased as total testosterone concentration increased in men with T2DM. After adjusting for age, current smoking, current drinking, body mass index, duration of T2DM, diastolic blood pressure, total cholesterol, triglycerides, low-density lipoprotein/high-density lipoprotein cholesterol ratio, uric acid, C-reactive protein, and sex hormones in model 4, the adjusted odds ratio (OR) and 95% confidence interval (CI) of NAFLD for tertile3 vs tertile1 was 0.37 (0.17-0.77; P = 0.024 for trend). When taken as a continuous variable, this association was still robust in model 4 (OR, 0.58; 95% CI, 0.42-0.80; P < 0.05). No significant associations were found between increasing levels of the precursors of androgens and the odds of NAFLD in men with T2DM (all P > 0.05). Moreover, women showed no significant associations of total testosterone, androstenedione, DHEA, and DHEAS, with the odds of NAFLD (all P > 0.05).

CONCLUSIONS

Serum total testosterone was independently associated with the risk of NAFLD among men with T2DM. This study highlights the potential role of testosterone as a risk factor for NAFLD in patients with T2DM.

Low Serum Dehydroepiandrosterone Is Associated With Diabetic Kidney Disease in Men With Type 2 Diabetes Mellitus

BACKGROUND

The associations of dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS) with diabetic kidney disease (DKD) remained unclear. Thus, this cross-sectional study aimed to explore the associations of DHEA and DHEAS with the risk of DKD in patients with T2DM.

METHODS

The information of 1251 patients with T2DM were included in this study. Serum DHEA and DHEAS were quantified using liquid chromatography-tandem mass spectrometry assays. Multivariate logistic regression analyses were used to assess the associations of DHEA and DHEAS with DKD as well as high urine albumin to creatinine ratio (ACR).

RESULTS

In men with T2DM, the risk of DKD decreased with an increasing DHEA concentration after adjustment for traditional risk factors; the fully adjusted OR (95% CI) for tertile3 vs tertile1 was 0.37 (0.19-0.70; P = 0.010 for trend). Similarly, when taking high ACR as the outcome, low DHEA levels were still significantly associated with increased odds of high ACR (OR, 0.37; 95% CI, 0.19-0.72 for tertile3 vs tertile1; P = 0.012 for trend). The restricted cubic spline showed that the risk of DKD gradually decreased with the increment of serum DHEA levels (P-overall = 0.007; P-nonlinear = 0.161). DHEAS was not independently associated with the risk of DKD in men. In contrast, no significant relationships were found between DHEA and DHEAS and the risk of DKD in women (all P > 0.05).

CONCLUSIONS

In men with T2DM, low serum DHEA levels were independently related to the risk of DKD after adjustment for traditional risk factors. Our finding highlights the potential role of DHEA in the development of DKD in men with T2DM.

Vitamin D Status and All-Cause Mortality in Patients With Type 2 Diabetes in China

OBJECTIVE

To assess the association between vitamin D status and all-cause mortality among type 2 diabetes patients.

RESEARCH DESIGN AND METHODS

We prospectively followed 1,291 participants with type 2 diabetes aged 20-80 years during 2013-2018. Cox proportional hazard regression models were used to estimate the association between different vitamin D status and all-cause mortality risk among hospitalized patients with type 2 diabetes.

RESULTS

During a median follow-up of 4.15 years (5,365 person-years in total), 61 cases of death were identified. Multivariable-adjusted hazard ratios (HRs) for all-cause mortality across the quartiles of baseline circulating 25-hydroxy vitamin D (25-OH vitamin D) were 2.70 [95% confidence interval (CI) 1.12-6.54], 1.00, 1.39 (95% CI 0.53-3.65), 2.31 (95% CI 0.96-5.54), respectively. Multivariable-adjusted HRs for all-cause mortality by different groups of baseline 25-OH vitamin D concentrations (<25, 25-49, 50-100, and ≥100 nmol/L) were 1.31 (95% CI 0.58-2.96), 0.94 (95% CI 0.47-1.87), 1.00, and 3.58 (95% CI 1.43-8.98), respectively.

CONCLUSIONS

Very low or high concentrations of vitamin D may be associated with a higher risk of all-cause mortality among patients with type 2 diabetes.

Criticality analysis for safety-critical software in nuclear power plant distributed control system

Software criticality analysis examines the degree of contribution that each individual failure mode of a software component has on the reliability of software. Higher safety integrity levels are assigned to software modules whose failures cause an unacceptable impact on the operation of the system, and these levels require the implementation of more rigorous software quality assurance measures as defined in IEEE Std 1012 and in the customer’s system requirements specification. In this paper, a novel software criticality analysis method is proposed, the results of which can be used to guide the development of newly developed software and the procurement of Commercial-Off-The-Shelf (COTS) software. The software structure is first analyzed and the software is divided into modules according to their functions. Then the criticality levels of software components are preliminarily classified by means of a safety criticality preliminary analysis tree, followed by their verification through the software hazard and operability analysis (HAZOP). Finally, the target Safety Integrity Level (SIL) of each software module is determined based on its criticality level and the overall safety objective (i. e., SIL) of the system it resides in. As an example, this proposed method is applied to a nuclear power plant safety-critical system to demonstrate the detail application process and to verify the feasibility of the method. Compared with the existing software criticality analysis methods, this method has better operability and verifiability, and can be utilized as a technical guidance for the software criticality analysis of nuclear power plant digital control systems.

Pancreatic neuroendocrine tumor producing vasopressin: A case report

RATIONALE

Functional pancreatic neuroendocrine tumors (pNETs) rarely produce vasopressin. Here, we reported a case of pNET producing vasopressin in a 78-year-old man with hyponatremia.

PATIENT CONCERNS

The patient presented with anorexia approximately 4 years ago, and the laboratory test results indicated hyponatremia. He was hospitalized 3 times subsequently due to anorexia in the past 4 years, during which laboratory tests consistently indicated severe hyponatremia.

DIAGNOSIS

Upon admission, his serum osmolarity, urine osmolarity, urine sodium level, and 24-hour urine sodium level was 277 mOsm/kg H2O, 465 mOsm/kg H2O, 82.5 mmol/L, and 140.25 mmol, respectively. Gallium-68-labeled tetraazacyclododecanetetraacetic acid-Dphel-Tyr3-octreotate positron emission tomography-computed tomography showed a high uptake lesion measuring approximately 1 cm in diameter in the pancreatic body, and the possibility of pNET was considered. Besides, laboratory tests showed that adrenocorticotropic hormone, follicle-stimulating hormone, and luteinizing hormone released by the pituitary was insufficient in the case of low levels of cortisol, estradiol, progesterone, and testosterone. Thus, the diagnosis of the syndrome of inappropriate antidiuresis (SIAD) was considered along with hypopituitarism.

INTERVENTIONS

The patient underwent surgery, and pNET was confirmed by pathology examination. The immunohistochemical study showed that the tumor cells were positive for somatostatin receptors 2 and vasopressin.

OUTCOMES

In the last follow-up 17 months after surgery, the patient was in good condition, taking methylprednisolone 4 mg every other day, and had been free of anorexia or hyponatremia episodes.

LESSONS

This case illustrated the potential ectopic production of vasopressin resulting in SIAD in pNETs, highlighting the adoption of gallium-68-labeled tetraazacyclododecanetetraacetic acid-Dphel-Tyr3-octreotate positron emission tomography-computed tomography and vasopressin immunohistochemical staining in the evaluation of the etiology of SIAD.

Development and validation of a novel index for the differential diagnosis of corticotropin-dependent Cushing syndrome

PURPOSE

To develop an index for the differential diagnosis of corticotropin-dependent Cushing syndrome (CS).

METHODS

The development cohort included 112 consecutive patients with clinicopathologically confirmed corticotropin-dependent CS at the Department of Endocrinology and Metabolism, Tianjin Medical University General Hospital, from December 2004 to May 2020, and data of 126 patients from studies published from 2016 to August 2020, identified through search in PubMed, Embase and the Cochrane Library, was extracted for external validation. The index was calculated as the product of plasma adrenocorticotropic hormone (ACTH, pmol/L) and urinary free cortisol (UFC, nmol/24 h) divided by 10,000. The discriminative ability was tested using receiver operating characteristics (ROC) curve analysis.

RESULTS

In development cohort, area under curve of ROC analysis of the ACTH-UFC index in identifying Cushing disease (CD) was 0.977. The diagnostic accuracy of ACTH-UFC index ≤ 11 was comparable to that of 48 h 8 mg/d high-dose dexamethasone test (HDDST) in identifying CD, with sensitivity, specificity, positive and negative likelihood ratios of 96.6%, 87.5%, 7.73, and 0.04, respectively. The sensitivity of ACTH-UFC index ≤ 11 in parallel combination with pituitary magnetic resonance imaging (MRI) was 100% for identifying CD. The performance of the ACTH-UFC index in parallel or serial combination with pituitary MRI was similar in the validation cohort.

CONCLUSIONS

ACTH-UFC index provides a rapid, convenient and non-invasive adjunctive approach for the differential diagnosis of corticotropin-dependent CS, with no risk of aggravating metabolic disturbances. Investigations for ectopic causes of corticotropin-dependent CS should be performed with ACTH-UFC index > 11 and negative contrasted pituitary MRI.

Relationship between onset of eclampsia and AGTR1 gene polymorphisms

OBJECTIVE

We aimed to study the correlations of angiotensin II receptor type 1 (AGTR1) gene polymorphisms with the occurrence and development of eclampsia.

PATIENTS AND METHODS

A total of 200 pregnant women with eclampsia admitted to our hospital from January 1, 2017 to September 30, 2019 were collected as observation group and 200 normal pregnant women during the same period were recruited in the control group. Genome sequencing was performed to detect the AGTR1 gene polymorphisms in the two groups. Expression level of AGTR1 in both groups was detected. The influences of AGTR1 on clinical data of pregnant women with eclampsia were analyzed.

RESULTS

There were no significant differences in age (p=0.545), height (p=0.738), weight (p=0.695) and hypertension (p=0.372) between observation group and control group. However, significant differences were found in the distributions of alleles at AGTR1 rs1799870 (p=0.002) and AGTR1 rs52936049 (p=0.047) between groups. The frequencies of T allele at rs1799870 and rs52936044 in observation group were higher than those in the control group. In addition, the distributions of AGTR1 gene genotypes at rs1799870 (p=0.012), rs144520513 (p=0.008) and rs529360494 (p p =0.036) in observation group differed from those in control group. Observation group had higher frequencies of TT genotype at rs1799870, GG genotype at rs144520513 and TG genotype at rs529360494 than those in control group. Besides, the frequency of CGG haplotype (p=0.008) of AGTR1 gene in observation group was notably lower than that in the control group, while the frequency of TGT haplotype (p=0.012) of AGTR1 gene in the former was remarkably higher than that in the latter. Moreover, the linkage disequilibrium between rs529360494 and rs144520513 of AGTR1 gene was relatively high (D'=0.623). AGTR1 gene polymorphism rs529360494 showed an evident relationship with the expression of AGTR1 gene, and the expression of AGTR1 in pregnant women with eclampsia who carried TG genotype was markedly reduced (p<0.05). Furthermore, AGTR1 gene polymorphism rs1799870 was associated with prothrombin time (PT) in pregnant women with eclampsia (p=0.046), and PT in those carrying genotype TC was shorter. Rs144520513 was related to the levels of triglyceride (TG) (p<0.001) and low-density lipoprotein (LDL) (p<0.001) in pregnant women with eclampsia, and TG and LDL levels were significantly lower.

CONCLUSIONS

AGTR1 gene polymorphisms are closely associated with the onset and progression of eclampsia.

Functional characterization of a loss-of-function mutant I324M of arginine vasopressin receptor 2 in X-linked nephrogenic diabetes insipidus

X-linked nephrogenic diabetes insipidus (X-linked NDI) is a rare inherited disease mainly caused by lost-of-function mutations in human AVPR2 gene encoding arginine vasopressin receptor 2 (V2R). Our focus of the current study is on exploration of the functional and biochemical properties of Ile324Met (I324M) mutation identified in a pedigree showing as typical recessive X-linked NDI. We demonstrated that I324M mutation interfered with the conformation of complex glycosylation of V2R. Moreover, almost all of the I324M-V2R failed to express on the cell surface due to being captured by the endoplasmic reticulum control system. We further examined the signaling activity of DDAVP-medicated cAMP and ERK1/2 pathways and the results revealed that the mutant receptor lost the ability in response to DDAVP stimulation contributed to the failure of accumulation of cAMP and phosphorylated ERK1/2. Based on the characteristics of molecular defects of I324M mutant, we selected two reagents (SR49059 and alvespimycin) to determine whether the functions of I324M-V2R can be restored and we found that both compounds can significantly “rescue” I324M mutation. Our findings may provide further insights for understanding the pathogenic mechanism of AVPR2 gene mutations and may offer some implications on development of promising treatments for patients with X-linked NDI.