T2-weighted imaging-based radiomic-clinical machine learning model for predicting the differentiation of colorectal adenocarcinoma

BACKGROUND The study on predicting the differentiation grade of colorectal cancer(CRC)based on magnetic resonance imaging(MRI)has not been reported yet.Developing a non-invasive model to predict the differentiation grade of CRC is of great value.AIM To develop and validate machine learning-based models for predicting the differ-entiation grade of CRC based on T2-weighted images(T2WI).METHODS We retrospectively collected the preoperative imaging and clinical data of 315 patients with CRC who underwent surgery from March 2018 to July 2023.Patients were randomly assigned to a training cohort(n=220)or a validation cohort(n=95)at a 7:3 ratio.Lesions were delineated layer by layer on high-resolution T2WI.Least absolute shrinkage and selection operator regression was applied to screen for radiomic features.Radiomics and clinical models were constructed using the multilayer perceptron(MLP)algorithm.These radiomic features and clinically relevant variables(selected based on a significance level of P<0.05 in the training set)were used to construct radiomics-clinical models.The performance of the three models(clinical,radiomic,and radiomic-clinical model)were evaluated using the area under the curve(AUC),calibration curve and decision curve analysis(DCA).RESULTS After feature selection,eight radiomic features were retained from the initial 1781 features to construct the radiomic model.Eight different classifiers,including logistic regression,support vector machine,k-nearest neighbours,random forest,extreme trees,extreme gradient boosting,light gradient boosting machine,and MLP,were used to construct the model,with MLP demonstrating the best diagnostic performance.The AUC of the radiomic-clinical model was 0.862(95%CI:0.796-0.927)in the training cohort and 0.761(95%CI:0.635-0.887)in the validation cohort.The AUC for the radiomic model was 0.796(95%CI:0.723-0.869)in the training cohort and 0.735(95%CI:0.604-0.866)in the validation cohort.The clinical model achieved an AUC of 0.751(95%CI:0.661-0.842)in the training cohort and 0.676(95%CI:0.525-0.827)in the validation cohort.All three models demonstrated good accuracy.In the training cohort,the AUC of the radiomic-clinical model was significantly greater than that of the clinical model(P=0.005)and the radiomic model(P=0.016).DCA confirmed the clinical practicality of incorporating radiomic features into the diagnostic process.CONCLUSION In this study,we successfully developed and validated a T2WI-based machine learning model as an auxiliary tool for the preoperative differentiation between well/moderately and poorly differentiated CRC.This novel approach may assist clinicians in personalizing treatment strategies for patients and improving treatment efficacy.

Comparison of Diagnostic Effects of T2-Weighted Imaging,DWI,SWI,and DTI in Acute Cerebral Infarction

Objective:To achieve precision medicine,the use of imaging methods to help the clinical detection of cerebral infarction is conducive to the clinical development of a treatment plan and increase of the cure rate and improvement of the prognosis of patients.Methods:In this work,T2-weighted imaging(T2WI),diffusion-weighted imaging(DWI),susceptibility-weighted imaging(SWI),and diffusion tensor imaging(DTI)examinations were performed on 34 patients with clinically diagnosed cerebral infarction to measure the difference in signal intensity between the lesion and its mirror area and make a comparative analysis by means of the Student-Newman-Keuls method.Results:The detection rate of T2WI was 79%(27/34),the detection rate of DWI was 97%(33/34),the detection rate of SWI was 88%(30/34),and the detection rate of DTI was 94%(32/34).Conclusion:The imaging performance was in the order DWI>DTI>SWI>T2WI for the diagnosis of cerebral infarction,and combined imaging is better than single imaging.

Early brainstem hemorrhage progression:multi-sequence magnetic resonance imaging and histopathology

According to clinical statistics,the mortality of patients with early brainstem hemorrhage is high.In this study,we established rat models of brainstem hemorrhage by injecting type Ⅶ collagenase into the right basotegmental pontine and investigated the pathological changes of early brainstem hemorrhage using multi-sequence magnetic resonance imaging and histopathological methods.We found that brainstem hematoma gradually formed in the injured rats over the first 3 days and then reduced after 7 days.The edema that occurred was mainly of the vasogenic type.No complete myelin sheath structure was found around the focus of the brainstem hemorrhage.The integrity and continuity of nerve fibers gradually deteriorated over the first 7 days.Neuronal degeneration was mild in the first 3 days and then obviously aggravated on the 7^(th)day.Inflammatory cytokines,interleukin-1β,and tumor necrosis factorαappeared on the 1st day after intracerebral hemorrhage,reached peak levels on the 3^(rd)day,and decreased from the 7^(th)day.Our findings show the characteristics of the progression of early brainstem hemorrhage.

A multiple-tissue-specific magnetic resonance imaging model for diagnosing Parkinson’s disease: a brain radiomics study

Brain radiomics can reflect the characteristics of brain pathophysiology.However,the value of T1-weighted images,quantitative susceptibility mapping,and R2*mapping in the diagnosis of Parkinson’s disease(PD)was underestimated in previous studies.In this prospective study to establish a model for PD diagnosis based on brain imaging information,we collected high-resolution T1-weighted images,R2*mapping,and quantitative susceptibility imaging data from 171 patients with PD and 179 healthy controls recruited from August 2014 to August 2019.According to the inclusion time,123 PD patients and 121 healthy controls were assigned to train the diagnostic model,while the remaining 106 subjects were assigned to the external validation dataset.We extracted 1408 radiomics features,and then used data-driven feature selection to identify informative features that were significant for discriminating patients with PD from normal controls on the training dataset.The informative features so identified were then used to construct a diagnostic model for PD.The constructed model contained 36 informative radiomics features,mainly representing abnormal subcortical iron distribution(especially in the substantia nigra),structural disorganization(e.g.,in the inferior temporal,paracentral,precuneus,insula,and precentral gyri),and texture misalignment in the subcortical nuclei(e.g.,caudate,globus pallidus,and thalamus).The predictive accuracy of the established model was 81.1±8.0%in the training dataset.On the external validation dataset,the established model showed predictive accuracy of 78.5±2.1%.In the tests of identifying early and drug-naïve PD patients from healthy controls,the accuracies of the model constructed on the same 36 informative features were 80.3±7.1%and 79.1±6.5%,respectively,while the accuracies were 80.4±6.3%and 82.9±5.8%for diagnosing middle-to-late PD and those receiving drug management,respectively.The accuracies for predicting tremor-dominant and non-tremor-dominant PD were 79.8±6.9%and 79.1±6.5%,respectively.In conclusion,the multiple-tissue-specific brain radiomics model constructed from magnetic resonance imaging has the ability to discriminate PD and exhibits the advantages for improving PD diagnosis.

Assessment of structural brain changes in patients with type 2 diabetes mellitus using the MRI-based brain atrophy and lesion index

Patients with type 2 diabetes mellitus(T2 DM) often have cognitive impairment and structural brain abnormalities.The magnetic resonance imaging(MRI)-based brain atrophy and lesion index can be used to evaluate common brain changes and their correlation with cognitive function,and can therefore also be used to reflect whole-brain structural changes related to T2 DM.A total of 136 participants(64 men and 72 women,aged 55–86 years) were recruited for our study between January 2014 and December 2016.All participants underwent MRI and Mini-Mental State Examination assessment(including 42 healthy control,38 T2 DM without cognitive impairment,26 with cognitive impairment but without T2 DM,and 30 T2 DM with cognitive impairment participants).The total and sub-category brain atrophy and lesion index scores in patients with T2 DM with cognitive impairment were higher than those in healthy controls.Differences in the brain atrophy and lesion index of gray matter lesions and subcortical dilated perivascular spaces were found between non-T2 DM patients with cognitive impairment and patients with T2 DM and cognitive impairment.After adjusting for age,the brain atrophy and lesion index retained its capacity to identify patients with T2 DM with cognitive impairment.These findings suggest that the brain atrophy and lesion index,based on T1-weighted and T2-weighted imaging,is of clinical value for identifying patients with T2 DM and cognitive impairment.Gray matter lesions and subcortical dilated perivascular spaces may be potential diagnostic markers of T2 DM that is complicated by cognitive impairment.This study was approved by the Medical Ethics Committee of University of South China(approval No.USC20131109003) on November 9,2013,and was retrospectively registered with the Chinese Clinical Trial Registry(registration No.Chi CTR1900024150) on June 27,2019.

Application of T2＊ measurement on gradient echo T2＊-weighted imaging in differential diagnosis of intracranial hemorrhage and calcification

Background Differential diagnosis of intracranial hemorrhage and calcification is a common problem encountered in clinical imaging diagnosis. The purpose of this study was to investigate the feasibility of T2＊ measurement on gradient echo （GRE） T2＊-weighted imaging （T2＊WI） in differential diagnosis of intracranial hemorrhage and calcification. Methods Thirty-eight hemorrhagic foci in 18 patients and 11 calcification foci in seven patients were included in this study. The diagnosis of hemorrhage and calcification was confirmed in all cases with enhanced T2＊ weighted angiography （ESWAN） magnetic resonance imaging （MRI） and CT respectively. The significance for the difference of T2＊ value between the central and peripheral areas of hemorrhage and calcification lesions was tested with univariate analysis of variance, Results The detection rate of GRE T2＊WI on intracranial hemorrhage was 1.9-fold higher than that of CT, especially for the hemorrhage in the brainstem and cerebellum. However, GRE T2＊WI was far less sensitive to calcification than CT. There was a significant difference in the T2＊ value between the central area of hemorrhage and calcification （P 〈0.001）, though no difference in the T2＊ value was obtained between the peripheral area of hemorrhage and calcification （P 〉0.05）. Conclusions Quantitative measurement of T2＊ value on GRE T2＊WI with a single MRI examination provides a fast, convenient, and effective means in differential diagnosis between intracranial hemorrhage and calcification, which may thus reduce the medical cost and save precious time for clinical management.

Using endogenous glycogen as relaxation agent for imaging liver metabolism by MRI

Glycogen plays essential roles in glucose metabolism.Imaging glycogen in the liver,the major glycogen reservoir in the body,may shed new light on many metabolic disorders.^(13)C magnetic resonance spectroscopy(MRS)has become the mainstream method for monitoring glycogen in the body.However,the equipment of special hard-ware to standard clinical magnetic resonance imaging(MRI)scanners limits its clinical applications.Herein,we utilized endogenous glycogen as a T_(2)-based relaxation contrast agent for imaging glycogen metabolism in the liver in vivo.The in vitro results demonstrated that the transverse relaxation rate of glycogen strongly correlates with the concentration,pH,and field strength.Based on the Swift-Connick theory,we characterized the exchange property of glycogen and measured the exchange rate of glycogen as 31,847 Hz at 37°C.Besides,the viscosity and echo spacing showed no apparent effect on the transverse relaxation rate.This unique feature enables vi-sualization of glycogen signaling in vivo through T_(2)-weighted MRI.Two hours-post intraperitoneal injection of glucagon,a clinical drug to promote glycogenolysis and gluconeogenesis,the signal intensity of the mice’s liver increased by 1.8 times from the T_(2)-weighted imaging experiment due to the decomposition of glycogen.This study provides a convenient imaging strategy to non-invasively investigate glycogen metabolism in the liver,which may find clinical applications in metabolic diseases.

CXCR4 Peptide Conjugated Au-Fe2O3 Nanoparticles for Tumor-targeting Magnetic Resonance Imaging

Peptide-functionalized Au-Fe2O3 nanoparticles（termed as anti-CXCR4-Au-Fe2O3 NPs） have been constructed through conjugation of dumbbell-like Au-Fe203 NPs with C-X-C motif chemokine receptor 4（CXCR4） binding cyclic peptide. One dumbbelMike Au-Fe2O3NP composes an Au NP[（3.3±0.3） nm in diameter] for conjugating CXCR4 binding cyclic peptide through Au-S covalent bond and a Fe2O3 NP[（8.7±0.8） nm in diameter] for using as T2-weighted magnetic resonance imaging（MRI） contrast agent. The anti-CXCR4-Au-Fe2O3 NPs have reasonable biocompatibility and integration of T2-weighted MRI contrast and tumor-targeting functionalities. The anti- CXCR4-Au-Fe2O3 NPs exhibit strong interactions with two kinds of breast tumor cells, MCF-7 cells and MDA-MB-231 cells, and high negative contrast in MRI of MDA-MB-231 tumor bearing mouse with 62% decreasing of MRI signal, indicating that the anti-CXCR4-Au-Fe2O3 NPs can recognize tumor with high efficacy and specificity.

腰椎间盘髓核退变的MRI研究(英文)

Objective The purpose of this study was to correlate T2 signal intensity values in nucleus pulposus of lumbar discs with patient age,gender and stage of nucleus pulposus degeneration.Methods Lumbar imaging findings of 422 cases subjects were retrospectively reviewed through T2 signal intensity values of nucleus pulposus evaluated based on the signal intensity values of MR T2-weighted mid-sagittal images of the lumbar spine,the t2 signal intensity values at all five lumbar levels (from L5-S1 to L1-L2) between male and female were used independent sample t-test and the Spearman correlation analysis.The age and grade of nucleus pulposus of disc degeneration and T2 signal intensity values were estimated by calculating and Chi-square test and the Spearman correlation analysis.The t-test was used to correlate the different anatomic levels of disc degeneration;T2 signal intensity values among the five different anatomic levels using non-parametric correlation analysis.Results There were significantly differences in T2 signal intensity values of nucleus pulposus at the same grade and anatomic level between male and female.Advanced with age,T2 signal intensity values of nucleus pulposus decreased and stage of disc degeneration increased accordingly.T2 signal intensity values may represent the nucleus pulposus degeneration of interverterbral disc.L4-L5 was the highest incidence among the nucleus pulposus degeneration of intervertebral disc.Conclusions The T2 signal intensity values based approach may be a feasible and practical tool to determine nucleus pulposus degeneration.T2 signal intensity values of nucleus pulposus of lumbar intervertebral disc are correlated with grade of degeneration and patient age.