Optical coherence tomography angiography for detection of microvascular changes in early diabetes: A systematic review and meta-analysis.

AIMS: To evaluate the effectiveness of optical coherence tomography angiography (OCTA) in detecting early intraocular microvascular changes in diabetic patients. MATERIALS AND METHODS: A systematic study search was performed on PubMed, Medline, Embase, and the Cochrane Library, ranging from January 2012 to March 2023. Controlled studies compared diabetes mellitus (DM) patients with non-diabetic retinopathy (NDR) or patients with mild non-proliferative diabetic retinopathy (mild NPDR) to healthy people. These studies included parameters of OCTA such as foveal avascular zone (FAZ), vessel density of superficial capillary plexus (VDscp), vessel density of deep capillary plexus (VDdcp), and peripapillary VD. The relevant effect model was used according to the heterogeneity, and the mean difference and 95% confidence intervals were calculated. RESULTS: A total of 18 studies with 2101 eyes were eventually included in this meta-analysis. Our results demonstrated that early alterations of VDscp, VDdcp, and peripapillary VD in NDR patients had a significant difference compared with healthy people by OCTA (VDscp: WMD = -1.34, 95% CI: -1.99 to -0.68, P < 0.0001. VDdcp: WMD = -2.00, 95% CI: -2.95 to -1.04, P < 0.0001. Peripapillary VD: WMD = -1.07, 95% CI: -1.70 to -0.43, P = 0.0010). However, there was no statistically significant difference in total FAZ between them (WMD = -0.00, 95% CI: -0.02-0.01, P = 0.84). In addition, for patients with mild NPDR, OCTA could illustrate prominent changes in VDscp, VDdcp, and total FAZ compared with healthy people (VDscp: WMD = -6.11, 95% CI: -9.90 to -2.32, P = 0.002. VDdcp: WMD = -4.26, 95% CI: -5.95 to -2.57, P < 0.00001. FAZ: WMD = 0.06, 95% CI: 0.01-0.11, P = 0.03). CONCLUSIONS: In diabetic patients with or without retinopathy, the parameters of OCTA such as VDscp, VDdcp, and peripapillary vessel density were demonstrated as potential biomarkers in monitoring the early alterations of retinal microangiopathy, while total FAZ may have no significant changes in diabetic patients without retinopathy.

AngioMT: A MATLAB based 2D image-to-physics tool to predict oxygen transport in vascularized microphysiological systems.

Microphysiological models (MPS) are increasingly getting recognized as in vitro preclinical systems of pathophysiology and drug discovery. However, there is also a growing need to adapt and advance MPS to include the physiological contributions of the capillary vascular dynamics, because they undergo angiogenesis or vasculogenesis to deliver soluble oxygen and nutrients to its organs. Currently, the process of formation of microvessels in MPS is measured arbitrarily, and vascularized MPS do not include oxygen measurements in their analysis. Sensing and measuring tissue oxygen delivery is extremely difficult because it requires access to opaque and deep tissue, and/or requires extensive integration of biosensors that makes such systems impractical to use in the real world. Here, a finite element method-based oxygen transport program, called AngioMT, is built in MATLAB. AngioMT processes the routinely acquired 2D confocal images of microvascular networks in vitro and solves physical equations of diffusion-reaction dominated oxygen transport phenomena. This user-friendly image-to-physics transition in AngioMT is an enabling tool of MPS analysis because unlike the averaged morphological measures of vessels, it provides information of the spatial transport of oxygen both within the microvessels and the surrounding tissue regions. Further, it solves the more complex higher order reaction mechanisms which also improve the physiological relevance of this tool when compared directly against in vivo measurements. Finally, the program is applied in a multicellular vascularized MPS by including the ability to define additional organ/tissue subtypes in complex co-cultured systems. Therefore, AngioMT serves as an analytical tool to enhance the predictive power and performance of MPS that incorporate microcirculation.

Association between retinal microvascular abnormalities and late-life brain amyloid-ß deposition: the ARIC-PET study.

BACKGROUND: Retinal microvascular signs are accessible measures of early alterations in microvascular dysregulation and have been associated with dementia; it is unclear if they are associated with AD (Alzheimer's disease) pathogenesis as a potential mechanistic link. This study aimed to test the association of retinal microvascular abnormalities in mid and late life and late life cerebral amyloid. METHODS: Participants from the ARIC-PET (Atherosclerosis Risk in Communities-Positron Emission Tomography) study with a valid retinal measure (N = 285) were included. The associations of mid- and late-life retinal signs with late-life amyloid-ß (Aß) by florbetapir PET were tested. Two different measures of Aß burden were included: (1) elevated amyloid (SUVR > 1.2) and (2) continuous amyloid SUVR. The retinal measures' association with Aß burden was assessed using logistic and robust linear regression models. A newly created retinal score, incorporating multiple markers of retinal abnormalities, was also evaluated in association with greater Aß burden. RESULTS: Retinopathy in midlife (OR (95% CI) = 0.36 (0.08, 1.40)) was not significantly associated with elevated amyloid burden. In late life, retinopathy was associated with increased continuous amyloid standardized value uptake ratio (SUVR) (ß (95%CI) = 0.16 (0.02, 0.32)) but not elevated amyloid burden (OR (95%CI) = 2.37 (0.66, 9.88)) when accounting for demographic, genetic and clinical risk factors. A high retinal score in late life, indicating a higher burden of retinal abnormalities, was also significantly associated with increased continuous amyloid SUVR (ß (95% CI) = 0.16 (0.04, 0.32)) independent of vascular risk factors. CONCLUSIONS: Retinopathy in late life may be an easily obtainable marker to help evaluate the mechanistic vascular pathway between retinal measures and dementia, perhaps acting via AD pathogenesis. Well-powered future studies with a greater number of retinal features and other microvascular signs are needed to test these findings.

Semi-automated protocol to quantify and characterize fluorescent three-dimensional vascular images.

The microvasculature facilitates gas exchange, provides nutrients to cells, and regulates blood flow in response to stimuli. Vascular abnormalities are an indicator of pathology for various conditions, such as compromised vessel integrity in small vessel disease and angiogenesis in tumors. Traditional immunohistochemistry enables the visualization of tissue cross-sections containing exogenously labeled vasculature. Although this approach can be utilized to quantify vascular changes within small fields of view, it is not a practical way to study the vasculature on the scale of whole organs. Three-dimensional (3D) imaging presents a more appropriate method to visualize the vascular architecture in tissue. Here we describe the complete protocol that we use to characterize the vasculature of different organs in mice encompassing the methods to fluorescently label vessels, optically clear tissue, collect 3D vascular images, and quantify these vascular images with a semi-automated approach. To validate the automated segmentation of vascular images, one user manually segmented one hundred random regions of interest across different vascular images. The automated segmentation results had an average sensitivity of 83±11% and an average specificity of 91±6% when compared to manual segmentation. Applying this procedure of image analysis presents a method to reliably quantify and characterize vascular networks in a timely fashion. This procedure is also applicable to other methods of tissue clearing and vascular labels that generate 3D images of microvasculature.

Revealing the complexity of meniscus microvasculature through 3D visualization and analysis.

Three-dimensional information is essential for a proper understanding of the healing potential of the menisci and their overall role in the knee joint. However, to date, the study of meniscal vascularity has relied primarily on two-dimensional imaging techniques. Here we present a method to elucidate the intricate 3D meniscal vascular network, revealing its spatial arrangement, connectivity and density. A polymerizing contrast agent was injected into the femoral artery of human cadaver legs, and the meniscal microvasculature was examined using micro-computed tomography at different levels of detail and resolution. The 3D vascular network was quantitatively assessed in a zone-base analysis using parameters such as diameter, length, tortuosity, and branching patterns. The results of this study revealed distinct vascular patterns within the meniscus, with the highest vascular volume found in the outer perimeniscal zone. Variations in vascular parameters were found between the different circumferential and radial meniscal zones. Moreover, through state-of-the-art 3D visualization using micro-CT, this study highlighted the importance of spatial resolution in accurately characterizing the vascular network. These findings, both from this study and from future research using this technique, improve our understanding of microvascular distribution, which may lead to improved therapeutic strategies.

Use superb microvascular imaging to diagnose and predict metastatic cervical lymph nodes in patients with papillary thyroid carcinoma.

PURPOSE: Papillary thyroid carcinoma (PTC) with metastatic lymph nodes (LNs) is closely associated with disease recurrence. This study accessed the value of superb microvascular imaging (SMI) in the diagnosis and prediction of metastatic cervical LNs in patients with PTC. METHODS: A total of 183 cervical LNs (103 metastatic and 80 reactive) from 116 patients with PTC were analysed. Metastatic cervical LNs were confirmed by pathology or/and cytology; reactive cervical LNs were confirmed by pathology or clinical features. The characteristic of conventional ultrasound (US) was extracted using univariate and multivariate analyses. The diagnostic performance of US and SMI were compared using the area under the receiver operating curve (AUC) with corresponding sensitivity and specificity. A nomogram was developed to predict metastatic LNs in patients with PTC, based on multivariate analyses. RESULTS: L/S < 2, ill-defined border, absence of hilum, isoechoic or hyperechoic, heterogeneous internal echo, peripheral or mixed vascular pattern on color Doppler flow imaging (CDFI) and SMI, and a larger SMI vascular index appeared more frequently in metastatic LNs in the training datasets than in reactive LNs (P < 0.05). The diagnostic sensitivity, specificity and accuracy of SMI vs US are 94.4% and 87.3%, 79.3% and 69.3%, and 87.6% and 79.1%, respectively; SMI combined with US exhibited a higher AUC [0.926 (0.877-0.975)] than US only [0.829 (0.759-0.900)]. L/S < 2, peripheral or mixed vascular type on CDFI, and peripheral or mixed vascular types on SMI were independent predictors of metastatic LNs with PTC. The nomogram based on these three parameters exhibited excellent discrimination, with an AUC of 0.926. CONCLUSION: SMI was superior to US in diagnosing metastatic LNs in PTC. US combined with SMI significantly improved the diagnostic accuracy of metastatic cervical LNs with PTC. SMI is efficacious for differentiating and predicting metastatic cervical LNs.

Speckle Variance Photoacoustic Microscopy for Microhemodynamic Imaging.

Relying on the strong optical absorption of hemoglobin to pulsed laser energy, photoacoustic microscopy provides morphological and functional information on microvasculature label-freely. Here, we propose speckle variance photoacoustic microscopy (SV-PAM), which harnesses intrinsic imaging contrast from temporal-varied photoacoustic signals of moving red blood cells in blood vessels, for recovering three-dimension hemodynamic images down to capillary-level resolution within the microcirculatory tissue beds in vivo. Calculating the speckle variance of consecutive photoacoustic B-scan frames acquired at the same lateral position enables accurate identification of blood perfusion and occlusion, which provides interpretations of dynamic blood flow in the microvasculature, in addition to the microvascular anatomic structures. We demonstrate high-resolution hemodynamic imaging of vascular occlusion and reperfusion in the microvasculature of mice ears in vivo. The results suggest that our SV-PAM is potentially invaluable for biomedical hemodynamic investigations, for example, imaging ischemic stroke and hemorrhagic stroke.

A mathematical model of tissue axial and radial diffusion in the microvasculature for intravascular microscopy and phosphorescence quenching data.

This study aims to extend earlier Krogh Cylinder Models of an oxygen profile by considering axial diffusion and analytically solving Fick's Law Partial Differential Equation with novel boundary conditions via the separation of variables. We next prospectively collected a total of 20 animals, which were randomly assigned to receive either fresh or two-week-old stored red blood cell (RBC) transfusions and PQM oxygen data were measured acutely (90 min) or chronically (24 h). Transfusion effects were evaluated in vivo using intravital microscopy of the dorsal skinfold window chamber in Golden Syrian Hamsters. Hamsters were initially hemorrhaged by 50% of total blood volume and resuscitated 1-h post hemorrhage. PQM data were subsequently collected and fit the derived 2D Krogh cylinder model. Systemic hemodynamics (mean arterial pressure, heart rate) were similar in both pre and post-transfusion with either stored or fresh cells. Transfusion with stored cells was found to impair axial and radial oxygen gradients as quantified by our model and consistent with previous studies. Specifically, we observed a statistically significant decrease in the arteriolar tissue radial oxygen gradient after transfusion with stored RBCs at 24 h compared with fresh RBCs (0.33 ± 0.17 mmHg µ m-1 vs, 0.14 ± 0.12 mmHg µ m-1; p = 0.0280). We also observed a deficit in the arteriolar tissue oxygen gradient (0.03 ± 0.01 mmHg µ m-1 fresh vs. 0.018 ± 0.007 mmHg µ m-1 stored; p = 0.0185). We successfully derived and validated an analytical 2D Krogh cylinder model in an animal model of microhemodynamic oxygen diffusion aberration secondary to storage lesions.

Context-aware deep learning enables high-efficacy localization of high concentration microbubbles for super-resolution ultrasound localization microscopy.

Ultrasound localization microscopy (ULM) enables deep tissue microvascular imaging by localizing and tracking intravenously injected microbubbles circulating in the bloodstream. However, conventional localization techniques require spatially isolated microbubbles, resulting in prolonged imaging time to obtain detailed microvascular maps. Here, we introduce LOcalization with Context Awareness (LOCA)-ULM, a deep learning-based microbubble simulation and localization pipeline designed to enhance localization performance in high microbubble concentrations. In silico, LOCA-ULM enhanced microbubble detection accuracy to 97.8% and reduced the missing rate to 23.8%, outperforming conventional and deep learning-based localization methods up to 17.4% in accuracy and 37.6% in missing rate reduction. In in vivo rat brain imaging, LOCA-ULM revealed dense cerebrovascular networks and spatially adjacent microvessels undetected by conventional ULM. We further demonstrate the superior localization performance of LOCA-ULM in functional ULM (fULM) where LOCA-ULM significantly increased the functional imaging sensitivity of fULM to hemodynamic responses invoked by whisker stimulations in the rat brain.

Long-term Intravital Investigation of an Orthotopic Glioma Mouse Model via Optical Coherence Tomography Angiography.

BACKGROUND/AIM: Probing brain tumor microvasculature holds significant importance in both basic cancer research and medical practice for tracking tumor development and assessing treatment outcomes. However, few imaging methods commonly used in clinics can noninvasively monitor the brain microvascular network at high precision and without exogenous contrast agents in vivo. The present study aimed to investigate the characteristics of microvasculature during brain tumor development in an orthotopic glioma mouse model. MATERIALS AND METHODS: An orthotopic glioma mouse model was established by surgical orthotopic implantation of U87-MG-luc cells into the mouse brain. Then, optical coherence tomography angiography (OCTA) was utilized to characterize the microvasculature progression within 14 days. RESULTS: The orthotopic glioma mouse model evaluated by bioluminescence imaging and MRI was successfully generated. As the tumor grew, the microvessels within the tumor area slowly decreased, progressing from the center to the periphery for 14 days. CONCLUSION: This study highlights the potential of OCTA as a useful tool to noninvasively visualize the brain microvascular network at high precision and without any exogenous contrast agents in vivo.

Microvascular Perfusion Imaging in Alzheimer's Disease.

Alzheimer's disease (AD) is the leading cause of dementia worldwide and significantly impacts the essential functions of daily life and social activities. Research on AD has found that its pathogenesis is related to the extracellular accumulation of amyloid-beta (Aß) plaques and intracellular neurofibrillary tangles in the cortical and limbic areas of the human brain, as well as cerebrovascular factors. The detection of Aß or tau can be performed using various probes and methodologies. However, these modalities are expensive to implement and often require invasive procedures, limiting accessibility on a large scale. While magnetic resonance imaging (MRI) and computed tomography (CT) are generally used for morphological and structural brain imaging, they show wide variability in their accuracy for the clinical diagnosis of AD. Several novel imaging modalities have emerged as alternatives that can accurately and vividly display the changes in blood flow and metabolism in each brain area and enable physicians and researchers to gain insights into the generation and progression of the cerebro-microvascular pathologies of AD. In this review, we summarize the current knowledge on microvascular perfusion imaging modalities and their application in AD, including MRI (dynamic susceptibility contrast-MRI, arterial spin labeling-MRI), CT (cerebral CT perfusion imaging), emission computed tomography (positron emission tomography (PET), single-photon emission computed tomography (SPECT)), transcranial doppler ultrasonography (TCD), and retinal microvascular imaging (optical coherence tomography imaging, computer-assisted methods for evaluating retinal vasculature).

Investigation of air bubble behaviour after gas embolism events induced in a microfluidic network mimicking microvasculature.

Gas embolism is a medical condition that occurs when gas bubbles are present in veins or arteries, decreasing blood flow and potentially reducing oxygen delivery to vital organs, such as the brain. Although usually reported as rare, gas embolism can lead to severe neurological damage or death. However, presently, only limited understanding exists regarding the microscale processes leading to the formation, persistence, movement, and resolution of gas emboli, as modulated by microvasculature geometrical features and blood properties. Because gas embolism is initially a physico-chemical-only process, with biological responses starting later, the opportunity exists to fully study the genesis and evolution of gas emboli using in vitro microfluidic networks mimicking small regions of microvasculature. The microfluidics networks used in this study, which aim to mimic microvasculature geometry, comprise linear channels with T-, or Y-junction air inlets, with 20, 40, and 60 µm widths (arterial or venous), and a 30 µm width honeycombed network (arterial) with three bifurcation angles (30°, 60°, and 90°). Synthetic blood, equivalent to 46% haematocrit concentrations, and water were used to study the modulation of gas embolism-like events by liquid viscosity. Our study shows that (i) longer bubbles with lower velocity occur in narrower channels, e.g., with 20 µm width; (ii) the resistance of air bubbles to the flow increases with the higher haematocrit concentration; and lastly (iii) the propensity of gas embolism-like events in honeycomb architectures increases for more acute, e.g., 30°, bifurcation angles. A dimensionless analysis using Euler, Weber, and capillary numbers demarcated the conditions conducive to gas embolism. This work suggests that in vitro experimentation using microfluidic devices with microvascular tissue-like structures could assist medical guidelines and management in preventing and mitigating the effects of gas embolism.

Coronary Microvascular Function Following Severe Preeclampsia.

BACKGROUND: Preeclampsia is a pregnancy-specific hypertensive disorder associated with an imbalance in circulating proangiogenic and antiangiogenic proteins. Preclinical evidence implicates microvascular dysfunction as a potential mediator of preeclampsia-associated cardiovascular risk. METHODS: Women with singleton pregnancies complicated by severe antepartum-onset preeclampsia and a comparator group with normotensive deliveries underwent cardiac positron emission tomography within 4 weeks of delivery. A control group of premenopausal, nonpostpartum women was also included. Myocardial flow reserve, myocardial blood flow, and coronary vascular resistance were compared across groups. sFlt-1 (soluble fms-like tyrosine kinase receptor-1) and PlGF (placental growth factor) were measured at imaging. RESULTS: The primary cohort included 19 women with severe preeclampsia (imaged at a mean of 15.3 days postpartum), 5 with normotensive pregnancy (mean, 14.4 days postpartum), and 13 nonpostpartum female controls. Preeclampsia was associated with lower myocardial flow reserve (ß, -0.67 [95% CI, -1.21 to -0.13]; P=0.016), lower stress myocardial blood flow (ß, -0.68 [95% CI, -1.07 to -0.29] mL/min per g; P=0.001), and higher stress coronary vascular resistance (ß, +12.4 [95% CI, 6.0 to 18.7] mm Hg/mL per min/g; P=0.001) versus nonpostpartum controls. Myocardial flow reserve and coronary vascular resistance after normotensive pregnancy were intermediate between preeclamptic and nonpostpartum groups. Following preeclampsia, myocardial flow reserve was positively associated with time following delivery (P=0.008). The sFlt-1/PlGF ratio strongly correlated with rest myocardial blood flow (r=0.71; P<0.001), independent of hemodynamics. CONCLUSIONS: In this exploratory cross-sectional study, we observed reduced coronary microvascular function in the early postpartum period following preeclampsia, suggesting that systemic microvascular dysfunction in preeclampsia involves coronary microcirculation. Further research is needed to establish interventions to mitigate the risk of preeclampsia-associated cardiovascular disease.

Prediction of complete response after neoadjuvant chemotherapy for invasive breast cancers: The utility of shear wave elastography and superb microvascular imaging in pretreatment breast ultrasound.

PURPOSE: To investigate whether multiparametric parameters of pretreatment breast ultrasound (US) and clinicopathologic factors are associated with pathologic complete response (pCR) after neoadjuvant chemotherapy (NAC) for breast cancer. METHODS: Between November 2018 and September 2022, 88 patients who underwent NAC and subsequent surgery were included in this study (median age, 55 years; interquartile range [IQR], 45, 59.3). Multiparametric breast US including grayscale, shear wave elastography (SWE) and superb microvascular imaging (SMI) of pathologically proven invasive breast cancers were retrospectively reviewed. Clinicopathological and multiparametric parameters of breast US, including size, SWEmax, SWEratio and vascular index on SMI (SMIVI) were compared between the groups. Univariate and multivariate logistic regression analyses were performed to determine factors predicting pCR after NAC. AUROC curve analysis was performed to determine the predictors' optimal cut-off values and diagnostic performance. RESULTS: The pCR group (n = 24) showed a significantly smaller tumor size, lower SWEmax, higher Ki-67 index, higher hormone receptor negativity and negative axillary lymph node metastasis compared to the non-pCR group (n = 64). Multivariate regression analysis showed that SWEmax (adjusted odds ratio[aOR] = 0.956, 95 % confidence interval [CI] = 0.919-0.994, P = 0.025) and Ki-67 index (aOR = 1.083, 95 % CI = 1.012-1.159, P = 0.021) were independently associated with pathologically complete response. The optimal cut-off values for predicting pCR were 27.5 % for Ki-67 with an AUC of 0.743 and 134.8 kPa for SWEmax with an AUC of 0.779. A combination model including clinical factors and SWEmax showed the best diagnostic performance with an AUC of 0.876. CONCLUSION: A higher Ki-67 index and lower SWEmax measured on pretreatment breast US were independently associated with pCR in invasive breast cancer after NAC.

Multitask deep learning for prediction of microvascular invasion and recurrence-free survival in hepatocellular carcinoma based on MRI images.

BACKGROUND AND AIMS: Accurate preoperative prediction of microvascular invasion (MVI) and recurrence-free survival (RFS) is vital for personalised hepatocellular carcinoma (HCC) management. We developed a multitask deep learning model to predict MVI and RFS using preoperative MRI scans. METHODS: Utilising a retrospective dataset of 725 HCC patients from seven institutions, we developed and validated a multitask deep learning model focused on predicting MVI and RFS. The model employs a transformer architecture to extract critical features from preoperative MRI scans. It was trained on a set of 234 patients and internally validated on a set of 58 patients. External validation was performed using three independent sets (n = 212, 111, 110). RESULTS: The multitask deep learning model yielded high MVI prediction accuracy, with AUC values of 0.918 for the training set and 0.800 for the internal test set. In external test sets, AUC values were 0.837, 0.815 and 0.800. Radiologists' sensitivity and inter-rater agreement for MVI prediction improved significantly when integrated with the model. For RFS, the model achieved C-index values of 0.763 in the training set and ranged between 0.628 and 0.728 in external test sets. Notably, PA-TACE improved RFS only in patients predicted to have high MVI risk and low survival scores (p < .001). CONCLUSIONS: Our deep learning model allows accurate MVI and survival prediction in HCC patients. Prospective studies are warranted to assess the clinical utility of this model in guiding personalised treatment in conjunction with clinical criteria.

Predicting microvascular invasion in hepatocellular carcinoma with a CT- and MRI-based multimodal deep learning model.

PURPOSE: To investigate the value of a multimodal deep learning (MDL) model based on computed tomography (CT) and magnetic resonance imaging (MRI) for predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC). METHODS: A total of 287 patients with HCC from our institution and 58 patients from another individual institution were included. Among these, 119 patients with only CT data and 116 patients with only MRI data were selected for single-modality deep learning model development, after which select parameters were migrated for MDL model development with transfer learning (TL). In addition, 110 patients with simultaneous CT and MRI data were divided into a training cohort (n = 66) and a validation cohort (n = 44). We input the features extracted from DenseNet121 into an extreme learning machine (ELM) classifier to construct a classification model. RESULTS: The area under the curve (AUC) of the MDL model was 0.844, which was superior to that of the single-phase CT (AUC = 0.706-0.776, P < 0.05), single-sequence MRI (AUC = 0.706-0.717, P < 0.05), single-modality DL model (AUCall-phase CT = 0.722, AUCall-sequence MRI = 0.731; P < 0.05), clinical (AUC = 0.648, P < 0.05), but not to that of the delay phase (DP) and in-phase (IP) MRI and portal venous phase (PVP) CT models. The MDL model achieved better performance than models described above (P < 0.05). When combined with clinical features, the AUC of the MDL model increased from 0.844 to 0.871. A nomogram, combining deep learning signatures (DLS) and clinical indicators for MDL models, demonstrated a greater overall net gain than the MDL models (P < 0.05). CONCLUSION: The MDL model is a valuable noninvasive technique for preoperatively predicting MVI in HCC.

Dual-exposure temporal laser speckle imaging for simultaneously accessing microvascular blood perfusion and angiography.

Laser speckle contrast imaging (LSCI) has gained significant attention in the biomedical field for its ability to map the spatio-temporal dynamics of blood perfusion in vivo. However, LSCI faces difficulties in accurately resolving blood perfusion in microvessels. Although the transmissive detecting geometry can improve the spatial resolution of tissue imaging, ballistic photons directly transmitting forward through tissue without scattering will cause misestimating in the flow speed by LSCI because of the lack of a quantitative theoretical model of transmissvie LSCI. Here, we develop a model of temporal LSCI which accounts for the effect of nonscattered light on estimating decorrelation time. Based on this model, we further propose a dual-exposure temporal laser speckle imaging method (dEtLSCI) to correct the overestimation of background speed when performing traditional transmissive LSCI, and reconstruct microvascular angiography using the scattered component extracted from total transmitted light. Experimental results demonstrated that our new method opens an opportunity for LSCI to simultaneously resolve the blood vessels morphology and blood flow speed at microvascular level in various contexts, ranging from the drug-induced vascular response to angiogenesis and the blood perfusion monitoring during tumor growth.

Clinical prediction of microvascular invasion in hepatocellular carcinoma using an MRI-based graph convolutional network model integrated with nomogram.

OBJECTIVES: Based on enhanced MRI, a prediction model of microvascular invasion (MVI) for hepatocellular carcinoma (HCC) was developed using graph convolutional network (GCN) combined nomogram. METHODS: We retrospectively collected 182 HCC patients confirmed histopathologically, all of them performed enhanced MRI before surgery. The patients were randomly divided into training and validation groups. Radiomics features were extracted from the arterial phase (AP), portal venous phase (PVP), and delayed phase (DP), respectively. After removing redundant features, the graph structure by constructing the distance matrix with the feature matrix was built. Screening the superior phases and acquired GCN Score (GS). Finally, combining clinical, radiological and GS established the predicting nomogram. RESULTS: 27.5% (50/182) patients were with MVI positive. In radiological analysis, intratumoural artery (P = 0.007) was an independent predictor of MVI. GCN model with grey-level cooccurrence matrix-grey-level run length matrix features exhibited area under the curves of the training group was 0.532, 0.690, and 0.885 and the validation group was 0.583, 0.580, and 0.854 for AP, PVP, and DP, respectively. DP was selected to develop final model and got GS. Combining GS with diameter, corona enhancement, mosaic architecture, and intratumoural artery constructed a nomogram which showed a C-index of 0.884 (95% CI: 0.829-0.927). CONCLUSIONS: The GCN model based on DP has a high predictive ability. A nomogram combining GS, clinical and radiological characteristics can be a simple and effective guiding tool for selecting HCC treatment options. ADVANCES IN KNOWLEDGE: GCN based on MRI could predict MVI on HCC.