Differential Diagnostic Model Research Articles

Identification of Alzheimer’s disease and vascular dementia based on a Deep Forest and near-infrared spectroscopy analysis method

Alzheimer’s disease (AD) and vascular dementia (VaD) typically do not exhibit distinct differences in clinical manifestations and auxiliary examination results, which leads to a high misdiagnosis rate. However, significant differences in treatment approaches and prognosis between these two diseases underscore the critical need for an accurate diagnosis of AD and VaD. In this study, serum samples from 33 patients with AD patients, 37 patients with VaD, and 130 healthy individuals were collected, employing near-infrared aquaphotomics technology in combination with deep learning for differential diagnoses. Through an analysis of water absorption patterns among different diseases via aquaphotomics, the efficacies of traditional machine learning methods (Support Vector Machine and Decision Trees) and deep learning approaches (Deep Forest) in modeling were compared. Ultimately, by leveraging feature extraction techniques in conjunction with deep learning, a differential diagnostic model for AD and VaD was successfully developed. The results revealed that aquaphotomics could identify a certain correlation between the number of hydrogen bonds in water molecules and the development of AD and VaD; the deep learning model was found to be superior to traditional machine learning models, achieving an accuracy of 98.67 %, sensitivity of 97.33 %, and specificity of 100.00 %. The bands identified using the Competitive Adaptive Reweighting Algorithm method, primarily located at approximately 1300–1500 nm, showed a significant correlation with water molecules containing four hydrogen bonds. These results highlighted the potential role of the water molecule hydrogen-bond network in disease development and were consistent with the aquaphotomics analysis results. Therefore, the differential diagnostic model developed by integrating near-infrared spectroscopy and deep learning was proven to be effective and feasible, providing accurate and rapid diagnostic methods for AD and VaD diagnoses.

The combined application of oligoclonal bands in cerebrospinal fluid and IgG intrathecal synthesis indicators and biochemical markers in the diagnosis of multiple sclerosis

Objective: To establish and verify a diagnostic model for distinguishing multiple sclerosis (MS) from other neurological diseases with similar symptoms by usingcerebrospinal fluid oligoclonal band (CSF-OCB)combined with IgG intrathecal synthesis indicators and biochemical markers. Methods: Multiple sclerosis (MS) patients admitted to the Neurology Department of Beijing Tiantan Hospital affiliated with Capital Medical University from January 2014 to December 2022 were selected as the case group, while patients with similar neurological symptoms were selected as the control group. Using the case-control study design, a retrospective analysis was conducted on the detection of age, gender, oligoclonal bands in cerebrospinal fluid, IgG intrathecal synthesis indicators and biochemical indicators for all study subjects. The differential diagnosis model was determined by the multiple logistic regression analysis, and the receiver operating characteristic (ROC) curve was used to analyze the diagnostic efficiency of the differential diagnosis model for neurological diseases with similar symptoms to MS and other conditions. Results: This study included 167 patients in the case group and 335 patients in the control group, of which 128 patients in the case group and 265 patients in the control group were used to construct the model, and 39 patients in the case group and 70 patients in the control group were used for model validation. The differential diagnostic model constructed by a multivariate logistic regression model was Y=0.871×CSF-OCB-0.051×CSFprotein-0.231×CSFchloride+1.183×gender-0.036×LDH+35.770. The model showed that the area under the curve, sensitivity and specificity were respectively 0.916, 87.3% and 87.6%. The Delong test results showed that the diagnostic efficacy of the model was significantly different from OCB, IgG intrathecal synthesis indicators, and OCB combined with IgG intrathecal synthesis indicators (P<0.05). The new model validation showed that the actual diagnostic consistency rate for the MS group was 84.6%, while the actual diagnostic consistency rate for the control group was 90.0%. Conclusion: This study combines OCB, IgG intrathecal synthesis indicators, and biochemical indicators to establish a diagnostic prediction model for neurological diseases with similar clinical symptoms in MS. This model may have good differential diagnostic value and can better assist clinical diagnosis in the early stages of disease progression in MS patients.

Early-stage diagnosis of ovarian cancer via digital immunoassay on a SlipChip

Ovarian cancer (OC) is one of the three major gynecologic malignancies and has the highest mortality rate because of the late diagnosis. Liquid biopsy based on serum protein biomarkers has demonstrated great potential for early diagnosis but remains limited by the analysis performance of conventional immunoassay technologies, such as chemiluminescence, and biomarkers, such as CA125. To address this challenge and achieve accurate early-stage diagnosis of OC, we developed a digital immunoassay on a SlipChip (DiSC) for quantitative analysis of a potential serum protein biomarker, Spondin-1 (SPON1). The DiSC system achieved a limit of detection (LoD) of 23 fg/μL for digital quantification of SPON1. The DiSC system was utilized to quantify the serum level of SPON1 in 357 clinical serum samples, including 63 from patients with benign ovarian tumors and 294 from patients with malignant ovarian cancer, ranging from stages I to IV. SPON1 concentrations were significantly different in samples from patients with malignant ovarian cancer. Notably, significantly different SPON1 levels were observed in early stages (I and II), in lymph node-negative cases (N0), and before metastasis (M0), suggesting that SPON1 could serve as a sensitive diagnostic biomarker for early-stage OC. The differential diagnostic model based on SPON1 levels quantified using DiSC demonstrated an area under the curve (AUC) of 0.8187 for early-stage OC, a significant improvement over CA-125 (AUC = 0.6967). For OC of all stages, the AUC was 0.8225, which could be further increased to 0.8750 when combined with CA-125. This showcases the potential of SPON1 as a novel biomarker for sensitive early-stage diagnosis of ovarian cancer and the capability of the DiSC system in discovering low-abundance biomarkers for disease diagnosis.

Correlation between retinal vascular geometric parameters and pathologically diagnosed type 2 diabetic nephropathy.

Diabetic nephropathy (DN) and diabetic retinopathy (DR) are common microvascular complications of diabetes. The purpose of this study was to investigate the correlation between retinal vascular geometric parameters and pathologically diagnosed type 2 DN and to determine the capacity of retinal vascular geometric parameters in differentiating DN from non-diabetic renal disease (NDRD). The study participants were adult patients with type 2 diabetes mellitus (T2DM) and chronic kidney disease who underwent a renal biopsy. Univariate and multivariable regression analyses were performed to evaluate associations between retinal vessel geometry parameters and pathologically diagnosed DN. Multivariate binary logistic regression analyses were performed to establish a differential diagnostic model for DN. In total, 403 patients were examined in this cross-sectional study, including 152 (37.7%) with DN, 157 (39.0%) with NDRD and 94 (23.3%) with DN combined with NDRD. After univariate logistic regression, total vessel fractal dimension, arteriolar fractal dimension and venular fractal dimension were all found to be associated with DN. In multivariate analyses adjusting for age, sex, blood pressure, diabetes, DR and other factors, smaller retinal vascular fractal dimensions were significantly associated with DN (P<.05). We developed a differential diagnostic model for DN combining traditional clinical indicators and retinal vascular geometric parameters. The area under the curve of the model established by multivariate logistic regression was 0.930. Retinal vessel fractal dimension is of great significance for the rapid and non-invasive differentiation of DN. Incorporating retinal vessel fractal dimension into the diagnostic model for DN and NDRD can improve the diagnostic efficiency.

Analysis of the expression patterns of AVP, IGF-1, and TNF-α, APP, CD44, IFN-β IFN A β-6, α-syn, and NFL and CLU genes in generalized and focal seizures

ObjectiveThe aim of our study was to investigate the relationship between clinical indicators and gene dysregulation in different types of epilepsy, while also seeking to identify a diagnostic model capable of distinguishing between focal and generalized seizures. This highlights the critical importance of understanding clinical indicators and gene dysregulation for targeted therapeutic interventions to effectively address the specific seizure types effectively. Materials and methodsIn this study, we conducted a comprehensive analysis of the peripheral blood of epilepsy patients (n = 100) and a control group (n = 51) to determine the differential gene expression. Our analysis involved a range of statistical approaches, including correlation analysis to establish the association between clinical indicators and gene dysregulation, and principal component analysis to highlight distinct disease group from control group. Furthermore, we developed diagnostic models using logistic regression to aid in the accurate diagnosis of epilepsy. ResultsAmong several selected genes in this study such as AVP (AUC = 0.832, p < 0.0001), IGF-1 (AUC = 0.658, p = 0.0015), TNF-α (AUC = 0.8970, p < 0.0001), APP (AUC = 0.742, p < 0.0001), CD44 (AUC = 0.614, p = 0.021) and NfL (AUC = 0.937, p < 0.0001), and CLU (AUC = 0.923, p < 0.0001) have shown the outstanding discrimination. In addition to this, when all genes were included in the model, the overall diagnostic power increased significantly (AUC = 0.9968). A differential diagnostic model for focal and generalized seizures was established which discloses AUC = 0.7027, (95 % CL, 0.5765 to 0.8289, p = 0.0019). ConclusionThe conclusions drawn from these findings represented that this is the first study to highlight the distinctive gene patterns of both focal and generalized seizures, implying that peripheral blood can serve as a diagnostic source to distinguish between these seizures types, aiding in the accurate classification of epilepsy. The findings from this study indicate a promising direction for investigating more targeted pharmacological interventions directed to address the distinct needs of both focal and generalized epilepsy, which offers advancements in treatment strategies for distinctive seizure types.

Construction of an early differentiation diagnosis model for patients with severe fever with thrombocytopenia syndrome and hemorrhagic fever with renal syndrome.

Severe fever with thrombocytopenia syndrome (SFTS) is an emerging infectious disease with a high mortality rate. Differentiating between SFTS and hemorrhagic fever with renal syndrome (HFRS) is difficult and inefficient. Retrospective analysis of the medical records of individuals with SFTS and HFRS was performed. Clinical and laboratory data were compared, and a diagnostic model was developed based on multivariate logistic regression analyzes. Receiver operating characteristic curve analysis was used to evaluate the diagnostic model. Among the 189 patients, 113 with SFTS and 76 with HFRS were enrolled. Univariate analysis revealed that more than 20 variables were significantly associated with SFTS. Multivariate logistic regression analysis revealed that gender, especially female gender (odds ratio [OR]: 4.299; 95% confidence interval [CI]: 1.163-15.887; p = 0.029), age ≥65 years (OR: 16.386; 95% CI: 3.043-88.245; p = 0.001), neurological symptoms (OR: 12.312; 95% CI: 1.638-92.530; p = 0.015), leukopenia (<4.0 × 109/L) (OR: 17.355; 95% CI: 3.920-76.839; p < 0.001), and normal Cr (OR: 97.678; 95% CI: 15.483-616.226; p < 0.001) were significantly associated with SFTS but not with HFRS. The area under the curve of the differential diagnostic model was 0.960 (95% CI: 0.936-0.984), which was significantly better than that of each single factor. In addition, the model exhibited very excellent sensitivity and specificity (92.9% and 85.5%, respectively). In cases where HFRS and SFTS are endemic, a diagnostic model based on five parameters, such as gender, age ≥65 years, neurological symptoms, leukopenia and normal Cr, will facilitate the differential diagnosis of SFTS and HFRS in medical institutions, especially in primary care settings.

A Case of Electroencephalography and Machine Learning in Early Diagnosis of Psychotic and Affective Disorders

AbstractElectroencephalography (EEG) serves as a non-invasive, cost-effective, and robust tool, directly measuring in-vivo neuronal mass activity with high temporal resolution. Using state-of-the-art machine learning techniques, EEG recordings have the potential to generate in silico biomarkers for severe mental disorders. In this study, we developed EEG-based classification models for schizophrenia and depression taking into account physiological and pathological aging processes. From a cohort (N=735, 51.6% male) that is acquired in LMU Hospital, Department of Psychiatry and Psychotherapy, comprising healthy control individuals (HC, N=245) and patients with schizophrenia (SCZ, N=250) or major depressive disorder (MDD, N=240), we extracted power spectrum density and connectivity measures based on 60 second resting-state EEG recordings with 19 channels. The support vector machine models were trained to 1) classify patients with SCZ or MDD and HC individuals, and 2) predict age in HC individuals using ten-by-ten repeated nested-cross validation. The age-predicting model was applied to patient groups to calculate EphysAGE (Electrophysiological Age Gap Estimation) by subtracting chronological age from chronological age. The links between diagnosis, medication, and EphysAGE, i.e., accelerated aging, were then further explored with univariate analyses. The EphysAGE Model had an explained variance of 46% (MAE=8.7 years, T=14.31, P1000&lt;0.001). The patients with SCZ had a significantly higher EphysAGE (mean[SD]=0.61[10.32]) than the patients with MDD (mean[SD]=-1.10[10.49], p=0.04). The classification models discriminated SCZ from HC (Balanced Accuracy, BAC=72.7%, p&lt;0.001), MDD from HC (BAC=67.0%, p&lt;0.001), and SCZ from MDD individuals (BAC=63.2%, p&lt;0.001). Higher EphysAGE was associated with an increased likelihood of being misclassified as SCZ in HC and MDD (ρHC=0.23, p&lt;0.001; ρMDD=0.17, p=0.01) based on percentile rank scores from the SCZ Model. Moreover, in the Differential Diagnostic Model, higher EphysAGE is positively correlated with being misclassified as SCZ in patients with MDD (ρMDD=0.14, p=0.03). Machine learning models can extract electrophysiological signatures of MDD and SCZ for potential clinical use. However, the impact of aging processes on diagnostic separability calls for timely application of such models, possibly in early recognition settings.Disclosure of InterestNone Declared

Mono+ algorithm assessment of the diagnostic value of dual-energy CT for high-risk factors for colorectal cancer: a preliminary study.

Risk factors for colorectal cancer (CRC) affect the way patients are subsequently treated and their prognosis. Dual-energy computerized tomography (DECT) is an advanced imaging technique that enables the quantitative evaluation of lesions. This study aimed to evaluate the quality of DECT images based on the Mono+ algorithm in CRC, and based on this, to assess the value of DECT in the diagnosis of CRC risk factors. This prospective study was performed from 2021 to 2023. A dual-phase DECT protocol was established for consecutive patients with primary CRC. The signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), overall image quality, lesion delineation, and image noise of the dual-phase DECT images were assessed. Next, the optimal energy-level image was selected to analyze the iodine concentration (IC), normalized iodine concentration (NIC), effective atomic number, electron density, dual-energy index (DEI), and slope of the energy spectrum curve within the tumor for the high- and low-risk CRC groups. A multifactor binary logistic regression analysis was used to construct a differential diagnostic regression model for high- and low-risk CRC, receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was calculated to assess the diagnostic value of the model. A total of 74 patients were enrolled in this study, of whom 41 had high-risk factors and 33 had low-risk factors. The SNR and CNR were best at 40 keV virtual monoenergetic imaging (VMI) based on the Mono+ algorithm (VMI+) (SNR 8.79±1.27, P<0.001; CNR 14.89±1.77, P=0.027). The overall image quality and lesion contours were best at 60 keV VMI+ and 40 keV VMI+, respectively (P=0.001). Among all the DECT parameters, the arterial phase (AP)-IC, NIC, DEI, energy spectrum curve, and venous phase-NIC differed significantly between the two groups. The AP-IC was the optimal DECT parameter for predicting high- and low-risk CRC with AUC, sensitivity, specificity, and cut-off values of 0.96, 97.06%, 87.80%, and 2.94, respectively, and the 95% confidence interval (CI) of the AUC was 0.88-0.99. Integrating the clinical factors and DECT parameters, the AUC, sensitivity, specificity, and predictive accuracy of the model were 0.99, 100.00%, 92.68%, and 94.67%, respectively, and the 95% CI of the AUC was 0.93-1.00. The DECT parameters based on 40 keV noise-optimized VMI+ reconstruction images depicted the CRC tumors best, and the clinical DECT model may have significant implications for the preoperative prediction of high-risk factors in CRC patients.

A hematological parameter-based model for distinguishing non-puerperal mastitis from invasive ductal carcinoma.

Non-puerperal mastitis (NPM) accounts for approximately 4-5% of all benign breast lesions. Ultrasound is the preferred method for screening breast diseases; however, similarities in imaging results can make it challenging to distinguish NPM from invasive ductal carcinoma (IDC). Our objective was to identify convenient and objective hematological markers to distinguish NPM from IDC. We recruited 89 patients with NPM, 88 with IDC, and 86 with fibroadenoma (FA), and compared their laboratory data at the time of admission. LASSO regression, univariate logistic regression, and multivariate logistic regression were used to screen the parameters for construction of diagnostic models. Receiver operating characteristic curves, calibration curves, and decision curves were constructed to evaluate the accuracy of this model. We found significant differences in routine laboratory data between patients with NPM and IDC, and these indicators were candidate biomarkers for distinguishing between the two diseases. Additionally, we evaluated the ability of some classic hematological markers reported in previous studies to differentiate between NPM and IDC, and the results showed that these indicators are not ideal biomarkers. Furthermore, through rigorous LASSO and logistic regression, we selected age, white blood cell count, and thrombin time to construct a differential diagnostic model that exhibited a high level of discrimination, with an area under the curve of 0.912 in the training set and with 0.851 in the validation set. Furthermore, using the same selection method, we constructed a differential diagnostic model for NPM and FA, which also demonstrated good performance with an area under the curve of 0.862 in the training set and with 0.854 in the validation set. Both of these two models achieved AUCs higher than the AUCs of models built using machine learning methods such as random forest, decision tree, and SVM in both the training and validation sets. Certain laboratory parameters on admission differed significantly between the NPM and IDC groups, and the constructed model was designated as a differential diagnostic marker. Our analysis showed that it has acceptable efficiency in distinguishing NPM from IDC and may be employed as an auxiliary diagnostic tool.

Serum cytokine profiles in patients with pancreatic cancer and chronic pancreatitis

BackgroundChronic pancreatitis (CP) may cause tumor-like lesions, creating a challenge in distinguishing between CP and pancreatic ductal adenocarcinoma (PDAC) in a patient. Given that invasive surgery is a standard cancer treatment, we aimed to examine whether a noninvasive diagnostic tool utilizing serum cytokines could safely differentiate between PDAC and CP. MethodsA pre-operative serum panel comprising 48 inflammatory cytokines, CA19-9, and C-reactive protein (CRP) was analyzed, consisting of 231 patients, 186 with stage I–III PDAC and 45 with CP. We excluded PDAC patients who underwent neoadjuvant therapy and those CP patients with other active malignancies. The laboratory variables most associated with PDAC diagnosis were assessed using logistic regression and selected using the lasso method. ResultsThe cytokines CTACK, GRO-α, and β-NGF were selected alongside CA19-9 and CRP for our differential diagnostic model. The area under the curve (AUC) for our differential diagnostic model was 0.809 (95% confidence interval [CI] 0.738–0.880), compared with 0.791 (95% CI 0.728–0.854) for CA19-9 alone (not significant). ConclusionsWe found that inflammatory cytokines CTACK, GRO-α, and β-NGF alongside CA19-9 and CRP may help distinguish PDAC from CP.

A pseudo-targeted metabolomics study based on serum bile acids profiling for the differential diagnosis of benign and malignant breast lesions

IntroductionBreast cancer (BC) has become the most commonly diagnosed cancer worldwide. It is very critical for the differential diagnosis between BC and benign breast diseases (BBD). The characteristics of serum bile acids (BAs) profiling in patients with BBD and BC was elucidated so that potential biomarkers could be find out for the differential diagnosis of BC and BBD. MethodsA pseudo-targeted approach was used to perform BAs metabolomics analysis in serum of 29 patients with BBD and 47 patients with BC by ultra-high performance liquid chromatography/hybrid quadrupole time-of-flight mass spectrometry (UPLC-QTOF-MS). Partial least squares-discriminant analysis (PLS-DA) was used to establish a differential diagnostic model for BC, and the receiver operating characteristic (ROC) curve and logistic regression analysis were used to screen out bile acids as biomarkers for the differential diagnosis of BC and BBD. ResultsThe serum BAs profile in BC group was quite different from that in BBD group. Compared with the BBD group, BC group had higher level of chenodeoxycholic acid (CDCA), while they had lower levels of dihydroxy tauro-conjugated BA (Tdi-1) and sulfated dihydroxy glyco-conjugated BA (Gdi-S-1). The sensitivity and specificity of PLS-DA model for patients classification were 100% and 92.3%, respectively. The combined biomarker, CDCA and Tdi-1, had high efficacy for the differential diagnosis (area under the curve was 0.954, 95% CI: 0.880-1.000) of BC. Besides, the performance was superior to traditional biomarkers in the differential diagnosis of BC with or without comorbidities. ConclusionThe profile of serum BAs in women with BC was quite different from that in patients with BBD. Serum BAs profiling analysis could be used as an effective tool for the differential diagnosis of BC and BBD.

A non-invasive differential diagnostic model for light chain cast nephropathy in newly diagnosed multiple myeloma patients with renal involvement: a multicenter study

Light chain cast nephropathy is the most common form of renal lesion in multiple myeloma. Kidney impairment caused by light chain cast nephropathy can be reversed and survival can be improved if early diagnosis is available. It is thus of imperative importance to develop a non-invasive method to diagnose light chain cast nephropathy once the kidney biopsy is not always applicable. We consecutively screened newly diagnosed multiple myeloma patients with kidney biopsies from 4 centers in China. Kidney pathologies were reviewed and clinical presentations were recorded. Then a diagnostic model was established by logistic regression and the predictive values were assessed. Between 1 June 1999 and 30 June 2019, a kidney biopsy was performed in 94 patients with newly diagnosed multiple myeloma, and light chain cast nephropathy was the most common pattern, seen in 52% of biopsied patients. The diagnostic model was established by multivariate logistic regression analysis as P(z) = 1/(1 + e-z) and z = -0.093 Hemoglobin (g/L) + 0.421 Serum albumin (g/L) + 3.463 Acute kidney injury (0/1) -9.207 High-density lipoprotein (mmol/L). If P(z) ≥ 0.55, the diagnosis pointed to light chain cast nephropathy; if P(z) < 0.55, the diagnosis favored non-light chain cast nephropathy. The area under the receiver operating characteristic curves was 0.981 (95% CI 0.959, 1.000). The model had a sensitivity of 93.9%, a specificity of 95.6%, a positive predictive value of 96.0%, a negative predictive value of 94.0%, and a total consistency of 95.0%. We built a novel, non-invasive diagnostic model through a multicenter study, which may be helpful in the diagnosis of light chain cast nephropathy in newly diagnosed multiple myeloma patients.

Laboratory indicators in COVID-19 and other pneumonias: Analysis for differential diagnosis and comparison of dynamic changes during 400-day follow-up

BackgroundCOVID-19 is spreading rapidly all over the world, the patients' symptoms can be easily confused with other pneumonia types. Therefore, it is valuable to seek a laboratory differential diagnostic protocol of COVID-19 and other pneumonia types on admission, and to compare the dynamic changes in laboratory indicators during follow-up. MethodsA total of 143 COVID-19, 143 bacterial pneumonia and 145 conventional viral pneumonia patients were included. The model group consisted of 140 COVID-19, 80 bacterial pneumonia and 60 conventional viral pneumonia patients, who were age and sex matched. We established a differential diagnostic model based on the laboratory results of the model group on admission via a nomogram, which was validated in an external validation group. We also compared the 400-day dynamic changes of the laboratory indicators among groups. ResultsLASSO regression and multivariate logistic regression showed that eosinophils (Eos), total protein (TP), prealbumin (PA), potassium (K), high-density lipoprotein cholesterol (HDLC), and low-density lipoprotein cholesterol (LDLC) could differentiate COVID-19 from other pneumonia types. The C-index of the nomogram model was 0.922. Applying the nomogram to the external validation group showed an area under the curve (AUC) of 0.902. The 400-day change trends of the laboratory indexes varied among subgroups divided by sex, age, oxygenation index (OI), and pathogen. ConclusionThe laboratory model was highly accurate at providing a new method to identify COVID-19 in pneumonia patients. The 400-day dynamic changes in laboratory indicators revealed that the recovery time of COVID-19 patients was not longer than that of other pneumonia types.

Easy-to-Use Scoring System Based on CT Pleural Fluid and Blood Features for Discriminating Malignant Pleural Effusions from Tuberculous Pleural Effusions: A Retrospective and Prospective Validation Study

Background: Distinguishing malignant pleural effusion (MPE) from tuberculous pleural effusion (TPE) is a challenge when symptom is atypical. Computed tomography (CT) scans and biochemical indicators are the most commonly used methods to evaluate pulmonary and pleural diseases, and the role of CT features (especially pleural and pleural fluid features) in the diagnosis of MPE or TPE is poorly described. Therefore, this study aims to establish and verify diagnostic models based on the characteristics of pleural and pleural effusion for the differential diagnosis of MPE and TPE. Methods: In this study, a first batch of 986 patients from Wuhan union hospital and Wuhan Jinyintan Hospital in the training set was used to develop differential diagnosis models. The CT features (including maximum pleural thickness at the non-mass/nodule, density of the pleural effusion, pleural mass/nodule, mediastinal pleural involvement, pleural wrap, pleural calcifications) and biochemical markers (including serum carcinoembryonic antigen (CEA), and adenosine deaminase (ADA) in the pleural fluid) were collected and subjected to logistic regression analyses. Then, two models were constructed: a model based on CT features (Model 1) and a model based on pleural fluid markers with or without blood biochemical markers (Models 2a/2b). The differential diagnostic model was displayed in the form of nomograms, which were validated by an independent prospective cohort (N=422). Findings: In the training set, eight variables were selected as significant risk factors for distinguishing MPE from TPE after multivariable regression analyses, i.e. age, sex, maximum pleural thickness at the non-nodule/mass, pleural fluid density, pleural nodule/mass, serum CEA, effusion CEA and effusion ADA. In Model 1, the area under the Receiver Operating Characteristic Curve (AUROC) was 0.970, yielding a sensitivity and specificity of 91.07% and 89.86% respectively in the training set, and AUROC = 0.968, sensitivity and specificity was 93.85% and 87.73% respectively in the validation set. In Model 2a, the AUROC, sensitivity and specificity for distinguishing MPE from TPE were 0.976, 92.73% and 93.66% in the training set, respectively, and 0.959, 93.50% and 87.84% in the validation set, respectively. In Model 2b, the AUROC, sensitivity and specificity for distinguishing MPE from TPE were 0.970, 94.87% and 90.87% in the training set, respectively, and 0.962, 96.77% and 80.25% in the validation set, respectively. There were no significant differences between the three models. Interpretation: CT features of pleura and pleural fluid could effectively help physicians differentiate between MPE and TPE, which were readily acquired and could be used as a non-invasive and convenient diagnostic tool in clinical practice. Trial Registration: This study was registered on the Clinical Trials website (No. NCT03997669). Funding Statement: This research was supported by the National Natural Science Foundation of China (No. 81770096; No. 81800094), the National Science and Technology Key Project (No. 2019ZX09301001). Declaration of Interests: The authors declare no competing non-financial/financial interests. Ethics Approval Statement: This study was approved by the Institutional Review Board of the Tongji Medical College, Huazhong University of Science and Technology, and supported by the Wuhan Clinical Research Center for Pleural Diseases

The novel potential biomarkers for multidrug-resistance tuberculosis using UPLC-Q-TOF-MS.

The lack of rapid and efficient diagnostics impedes largely the epidemic control of multidrug-resistant tuberculosis, and might misguide the therapeutic strategies as well. This study aimed to identify novel multidrug-resistant tuberculosis biomarkers to improve the early intervention, symptomatic treatment and control of the prevalence of multidrug-resistant tuberculosis. The serum small molecule metabolites in healthy controls, patients with drug-susceptible tuberculosis, and patients with multidrug-resistant tuberculosis were screened using ultra-high-performance liquid chromatography combined with quadrupole-time-of-flight mass spectrometry (UPLC-Q-TOF-MS). The differentially abundant metabolites were filtered out through multidimensional statistical analysis and bioinformatics analysis. Compared with drug-susceptible tuberculosis patients and healthy controls, the levels of 13 metabolites in multidrug-resistant tuberculosis patients altered. Among them, the most significant changes were found in N1-Methyl-2-pyridone-5-carboxamide (N1M2P5C), 1-Myristoyl-sn-glycerol-3-phosphocholine (MG3P), Caprylic acid (CA), and D-Xylulose (DX). And a multidrug-resistant tuberculosis/drug-susceptible tuberculosis differential diagnostic model was built based on these four metabolites, achieved the accuracy, sensitivity, and specificity of 0.928, 86.7%, and 86.7%, respectively. The enrichment analysis of metabolic pathways showed that the phospholipid remodeling of cell membranes was active in multidrug-resistant tuberculosis patients. In addition, in patients with tuberculosis, the metabolites of dipalmitoyl phosphatidylcholine (DPPC), a major component of pulmonary surfactant, were down-regulated. N1M2P5C, MG3P, CA, and DX may have the potential to serve as novel multidrug-resistant tuberculosis biomarkers. This research provides a preliminary experimental basis to further investigate potential multidrug-resistant tuberculosis biomarkers. Impact statement The MDR-TB incidence remains high, making the effective control of TB epidemic yet challenging. Rapid and accurate diagnosis is vitally important for improving the therapeutic efficacy and controlling the prevalence of drug resistance TB. Metabolomics has dramatic potential to distinguish MDR-TB and DS-TB. N1M2P5C, MG3P, CA, and DX that we identified in this study might have potential as novel MDR-TB biomarkers. The phospholipid remodeling of cell membranes was highly active in MDR-TB. The DPPC metabolites in TB were significantly down-regulated. This work aimed to investigate potential MDR-TB biomarkers to enhance the clinical diagnostic efficacy. The metabolic pathway distinctly altered in MDR-TB might provide novel targets to develop new anti-TB drugs.

Differential diagnosis of hematogenous osteomyelitis and malignant bone tumors

The objectiveof the study was to develop a mathematical model of differential diagnosis of hematogenous osteomyelitis and malignant bone tumor.Material and methods. In this research we performed a retrospective analysis of the data on 127 patients with differential diagnosis between hematogenous osteomyelitis and malignant bone tumors. Retrospective validity indicators: sensitivity (Se) – 53.12 % (42.66–63.39 %), specificity (Sp) – 70.97 % (51.96–85.78 %). Data on 96 patients with subsequently diagnosed hematogenous osteomyelitis was compared with data on 31 patients with morphologically confirmed malignant bone neoplasms. In the research we.Results.The database was created based on studying the data of the retrospective group. It became the basis for the development of a mathematical model for the differential diagnosis of hematogenous osteomyelitis and malignant bone tumors. 13 prognostic criteria were identified, each of which is quantified. These criteria included the age of the patient, sex, the data on concomitant pathology, information about the localization of the pathological process, the indices of some laboratory studies, etc. An algorithm for the differential diagnosis of hematogenous.Conclusion.Our differential diagnostic model was tested in the prospective study (63 cases). Indicators of the validity of the test was: Sensitivity – 83.05 %, Specificity (Sp) – 86.67 %. Positive Predictive Value (PPV) was 96.08 %, Negative Predictive Value (NPV) – 56.52 %, Accuracy – 83.78 %, which allowed to shorten the time of examination and send patients to specialized hospitals.

Dermoscopy of chondrodermatitis nodularis helicis.

Chondrodermatitis nodularis helicis (CNH) is a benign auricular disease whose differentiation with nonpigmented tumors is mandatory. Clinical characteristics of CNH are well known, but there is no information about the dermoscopic features that could help differentiate CNH from squamous cell carcinoma and other non-melanoma skin cancers. To describe the dermoscopic appearance of CNH and to formulate a differential diagnostic model, we conducted a retrospective, single center, observational dermoscopic study on a sample of 189 biopsy-proven lesions: 25 CNH; 26 squamous cell carcinomas; 62 basal cell carcinomas and 76 other benign and malignant tumors. Univariate and multivariate analyses were conducted by logistic regression. The most significant dermoscopic finding for CNH was a peculiar global configuration (daisy pattern), consisting of white thick lines, radially arranged, converging to a central rounded yellow/brown clod (an erosion covered by keratin or sero-crust). This pattern achieved 92 and 98% of specificity for discriminating CNH with squamous cell carcinoma and basal cell carcinoma, respectively. In conclusion, dermoscopy is valuable for the diagnosis of CNH as a first screening tool because of a consistent global dermoscopic configuration (daisy pattern), consisting of radially arranged white thick lines surrounding a central rounded yellow/brown clod.

APPLICATION OF ARTIFICIAL NEURAL NETWORKS FOR DIFFERENTIAL DIAGNOSIS AND PROPHYLAXIS OF THE DISEASES OF HEPATOPANCREATODUODENAL ZONE

The purpose of the study was to evaluate the possibilities of neuronet differential diagnosis of hepatopancreatoduodenal zone diseases and their potential for the formation of an individualized preventive strategy. The work was performed on the basis of materials of 488 patients with the pathology of the hepatopancreatoduodenal zone. Neural network analysis of information on risk factors (sex, age, eating habits, stress, family status, bad habits) was used. Results. The sensitivity of the differential diagnostic model reaches Se = 84,7, m = 1,66 for peptic ulcer, Se = 81,4, m = 1,8 for pancreatitis and Se = 92,1, m = 1,24 for cholecystitis. Specificity levels equaled respectively to Sp = 91,5, m = 1,29, Sp = 90,1, m = 1,38 and Sp = 82,7, m = 1,75. The article also presents a method of developing an individualized diagnostic strategy using the artificial neural network.

СТВОРЕННЯ ПРОГРАМНОГО ІНСТРУМЕНТУ ДЛЯ ПРОВЕДЕННЯ ДИФЕРЕНЦІЙНОЇ ДІАГНОСТИКИ НЕВРОЛОГІЧНИХ ХВОРОБ МЕТОДОМ ПОТЕНЦІЙНИХ ФУНКЦІЙ

Informatization in medicine offers many opportunities to enhance quality of medical support, accuracy of diagnosis and provides usage of accumulated experience. Modern program systems are represented as additional tools to get appropriate advice.This article offers the way to provide help for neurology department doctor of NCSH "OKHMATDYT" during diagnosis determining. Decision has been made to design the program system for this purpose based on differential diagnostic model.The key problems in differential diagnosis are symptoms similarity between each other in one disease group and the absence of key symptom. Thus, we need the differential diagnostic model. It is built using the potential function method in characteristics space. Such space of characteristics is formed by 26 points - patients with their symptoms.The main feature of this method is decision function, building during recognition step united with learning, which became possible with modern powerful computers.

A Computer Based Model in Comparison with Sonography Imaging to Diagnosis of Acute Appendicitis in Iran.

Acute appendicitis overlaps with conditions of other diseases in terms of Symptoms and signs in the first hours of presentation. Ultrasound imaging and laboratory tests are usually used to decrease the diagnosis errors in the case of abdominal pain. However, same results may be happened using the mentioned examination tools for a string of diseases with abdominal pain. Moreover, those tests raise the medical costs for hospitals and patients. Clinical Decision Support Systems (CDSSs) can be used to assist the physicians to make the proper health care decisions particularly in the unreliable conditions. To improve the decision making process by physicians in diagnosis of acute appendicitis, an optimizing model was developed. The main objective is to discover a diagnostic model using the minimum clinical factors available in the first hours of abdominal pain. Fuzzy-rule based classifier is a known technique in the Decision Support Systems (DSSs). In this article thus the useful clinical factors were explored and the diagnosis knowledge was discovered using Honey Bee Reproduction Cycle (HRBC) algorithm in the Fuzzy-rule based system. In this model, the proposed algorithm created the Fuzzy rules as the diagnosis knowledge in an optimizing process. To evaluate the accuracy of the proposed model for diagnosing of appendicitis, a collection of data was gathered from abdominal patients who referred to the educational general hospitals in Ahvaz, Iran in 2014 to 2015 years. In this process, the proposed model was optimized first in a training phase using a training dataset, and then it was tested with the testing dataset. Then, the achieved results from the computer base model were compared with ultrasound imaging findings before surgery as well as other detection methods in the previous studies. The comparison results illustrated that the proposed hybrid classification model as a CDSS improves considerably the accuracy of acute appendicitis diagnosis. Experimental outcomes illustrated that the proposed algorithm improves considerably the optimization performance in the diagnostic problem with the accuracy rate of 89.9%. The mentioned rate was achieved while a limited range of factors as the input parameters were used in the hybrid model. The proposed differential diagnostic model can be used as a CDSS especially conditions in which access to costly equipment such as CT scans and Sonography tools are limited. The developed model improves the diagnosis time as well as the treatment costs for the patients with acute abdomen suspicious of acute appendicitis.