Comparison of Red Blood Cell (RBC) Histogram and Indices on Peripheral Blood Smears for the Detection of Anemia: A Study Protocol.

Background Anemia is characterized by a lower red blood cell (RBC) count than the usual rate or the average count that should be found in the blood. It is described by the reduction in the concentration of hemoglobin, the number of RBCs, and the O2-carrying capacity of the blood. A complete blood count (CBC) gives serious information on changes in the size and shape of RBCs and an indication of inclusion bodies that will help exclude anemia. RBC indices suggest potential reasons for the anemia in a particular patient. Thus, RBC indices can give some clues about the possible cause of anemia in a given patient with the disease. It is also necessary to look at the peripheral blood smear (PBS) and RBC indices and histogram to evaluate anemia adequately. Method The primary complaints of volunteers, such as clinical anemia as well as essential family medical history, will be the criteria for selection. After that, the main procedures that will be performed are the PBS and CBC. The blood sample will be drawn while the individual sits securely and in an aseptic setting. An ethylenediaminetetraacetic acid (EDTA) tube will be used to collect 2 ml of venous blood, which will then be gently mixed and processed on a five-part automated hematology analyzer (Coulter) for CBC test and PBS. Expected results The diagnostic capabilities of the PBS procedure are expected to match with RBC histograms and indices in the CBC report for the detection of anemia. Comparing both results with the PBS evaluated at the microscopic assessment, it is expected to likely provide a more precise and quantifiable result on RBC morphology and indices with greater accuracy for anemia detection in accordance with the PBS report. Implementing RBC histogram analysis in routine clinical practice is expected to enhance the diagnostic processes for anemia, leading to a more effective and timelier patient treatment. Conclusion These findings will suggest that applying RBC histograms and indices in CBC reports in supportive diagnosis with PBS analysis in the everyday clinical practice of hematology may improve the diagnoses and overall care of patients with anemia. Advanced research should confirm the efficacy of integrated approaches to increase diagnostic accuracy.

Assessment of histogram analysis in distinguishing between low-grade and high-grade brainstem glioma.

Objectives: This retrospective study was designed to evaluate the value of histogram analysis on the apparent diffusion coefficient (ADC) map in distinguishing between low-grade and high-grade brainstem glioma (BSG). In this article, we used the two VOI (volume of interest) placements of the entire tumour method and the entire solid part method, thus aiming to compare the diagnostic value between these two performances. Methods: A total of 28 patients (8 low-grade BSGs and 20 high-grade BSGs) with histological diagnosis of BSG. All victims underwent contrast-enhanced magnetic resonance imaging (MRI). We measured ADC histogram parameters (mean, median, SD, max, min, Kurtosis, Skewness, Entropy, Uniformity, and Variance) and calculated the ratios between tumour and normal brain parenchyma parameters in two methods. Independent samples test, Mann-Whitney U test, and ROC curve were used to determine each value's cut-off point, sensitivity, and specificity. Results: Among the method of VOI placing the entire tumour, the values of ADC_min, rADC_mean, rADC_median, and rADC_min are significantly different between these two neoplasms with cut-off values (sensitivity %, specificity %) of 776 x10-6 m2/s (62.5%, 90%), 2.1765 (62.5%, 95%), 2.1588 (50%, 100%), 1.0535 (100%, 50%), respectively. On the other hand, the method of VOI placing the entire solid part of the tumour showed significantly different in ADC_mean, ADC_median, ADC_min, rADC_mean, rADC_median, rADC_min at the cut-off values (sensitivity%, specificity %) of 1491 x10-6 m2/s (62.5%, 95%), 1438.9 x10-6 m2/s (62.5%, 90%), 862.5 x10-6 m2/s (75%, 100%), 2,112 (62.5%, 95%), 1.9748 (62.5%, 90%), 1.3735 (87.5%, 90%), respectively. Conclusions: The ADC histogram analysis is a promising approach to distinguishing low-grade BSG and high-grade BSG. The entire solid part VOI placement has a superior value compared to the whole tumour VOI placement. The rADC_mean showed the best performance in differentiating between these two entities, followed by ADC_min, rADC_mean, rADC_median, ADC_mean, and ADC_median.

Predicting the Ki-67 proliferation index in cervical cancer: a preliminary comparative study of four non-Gaussian diffusion-weighted imaging models combined with histogram analysis.

Background: The prognosis for patients with cervical cancer (CC) is strongly correlated with the Ki-67 proliferation index (PI). However, the Ki-67 PI obtained through biopsy has certain limitations. The non-Gaussian distribution diffusion model of magnetic resonance imaging (MRI) may play an important role in characterizing tissue heterogeneity. At present, there are limited data available concerning the prediction of Ki-67 PI using models based on histogram features of non-Gaussian diffusion distribution. This study aimed to determine whether preoperative histogram features from multiple non-Gaussian models of diffusion-weighted imaging can predict the Ki-67 PI in patients with CC. Methods: Our cross-sectional prospective study recruited a total of 53 patients suspected of having CC who underwent 3.0-T MRI at Sun Yat-sen Memorial Hospital of Sun Yat-sen University between January 2022 and January 2023. Fifteen b values (0-4,000 s/mm2) were used for diffusion-weighted imaging. A total of nine parameters from four non-Gaussian diffusion-weighted imaging models, including continuous-time random walk (CTRW), diffusion kurtosis imaging (DKI), fractional order calculus (FROC), and intravoxel incoherent motion (IVIM), were used. Whole-tumor volumetric histogram analysis of these parameters was then obtained. In logistic regression, significant histogram characteristics were statistically examined across two groups to build the final prediction model. To assess diagnostic parameters of the proposed model in the diagnosis of the Ki-67 PI, along with the sensitivity, specificity, and diagnostic accuracy of these various parameters from the four models, receiver operating feature analysis was applied. Results: Among the 53 patients (55.3±9.6 years, ranging from 23 to 79 years) included in the study, 15 had a Ki-67 PI ≤50% and 38 had a Ki-67 PI >50%. Univariable analysis determined that 12 histogram features were statistically different between the two groups. In multivariable logistic regression, we ultimately selected 6 histogram features to construct the final prediction model, with CTRW_α_10th percentile [odds ratio (OR) =0.955; 95% confidence interval (CI): 0.92-0.99; P=0.019], CTRW_α_robust mean absolute deviation (OR =0.893; 95% CI: 0.81-0.99; P=0.028), and CTRW_α_uniformity (OR =0.000, 95% CI: 0.00-0.90, P=0.047) being the independent predictive variables. The area under the curve of the combined prediction model was 0.845 (95% CI: 0.74-0.95), with a sensitivity of 78.9% (95% CI: 0.63-0.90), a specificity of 86.7% (95% CI: 0.60-0.98), an accuracy of 81.1% (95% CI: 0.68-0.91), a positive predictive value of 93.8% (95% CI: 0.79-0.99), and a negative predictive value of 61.9% (95% CI: 0.38-0.82). Conclusions: The histogram features of multiple non-Gaussian diffusion-weighted imaging can help to predict the Ki-67 PI of CC, providing a new method for the noninvasive evaluation of critical biological features of CC.

Combination of intravoxel incoherent motion histogram parameters and clinical characteristics for predicting response to neoadjuvant chemoradiation in patients with locally advanced rectal cancer.

OBJECTIVE: To explore the value of histogram parameters derived from intravoxel incoherent motion (IVIM) for predicting response to neoadjuvant chemoradiation (nCRT) in patients with locally advanced rectal cancer (LARC). METHODS: A total of 112 patients diagnosed with LARC who underwent IVIM-DWI prior to nCRT were enrolled in this study. The true diffusion coefficient (D), pseudo-diffusion coefficient (D*), and microvascular volume fraction (f) calculated from IVIM were recorded along with the histogram parameters. The patients were classified into the pathological complete response (pCR) group and the non-pCR group according to the tumor regression grade (TRG) system. Additionally, the patients were divided into low T stage (yp T0-2) and high T stage (ypT3-4) according to the pathologic T stage (ypT stage). Univariate logistic regression analysis was implemented to identify independent risk factors, including both clinical characteristics and IVIM histogram parameters. Subsequently, models for Clinical, Histogram, and Combined Clinical and Histogram were constructed using multivariable binary logistic regression analysis for the purpose of predicting pCR. The area under the receiver operating characteristic (ROC) curve (AUCs) was employed to evaluate the diagnostic performance of the three models. RESULTS: The values of D_ kurtosis, f_mean, and f_ median were significantly higher in the pCR group compared with the non-pCR group (all P < 0.05). The value of D*_ entropy was significantly lower in the pCR group compared with the non-pCR group (P < 0.05). The values of D_ kurtosis, f_mean, and f_ median were significantly higher in the low T stage group compared with the high T stage group (all P < 0.05). The value of D*_ entropy was significantly lower in the low T stage group compared with the high T stage group (P < 0.05). The ROC curves indicated that the Combined Clinical and Histogram model exhibited the best diagnostic performance in predicting the pCR patients with AUCs, sensitivity, specificity, and accuracy of 0.916, 83.33%, 85.23%, and 84.82%. CONCLUSIONS: The histogram parameters derived from IVIM have the potential to identify patients who have achieved pCR. Moreover, the combination of IVIM histogram parameters and clinical characteristics enhanced the diagnostic performance of IVIM histogram parameters.

Algorithms to reduce the edge effect and improve the flood histogram quality of a PET detector consisting of two pixelated crystal arrays.

PURPOSE: The performance of detectors is key for a PET scanner to achieve high spatial resolution and high sensitivity. This work aims to develop flood histogram generating algorithms to reduce the edge effect and improve the crystal identification of a PET detector consisting of two optically coupled pixelated scintillator detectors. METHODS: The PET detector consists of two optically coupled detectors, each consisting of a 23×23 LYSO crystal array with a crystal size of 1.0×1.0×20 mm3 read out by an 8×8 SiPM array with a pixel size of 3.0×3.0 mm2. The SiPM array is read out with a resistor network circuit to obtain four position encoding energy signals. A novel center of gravity (COG) positioning algorithm using six signals from the two detectors was proposed and compared to the traditional COG algorithms using either four or eight signals from the detectors. The raised-to-the-power (RTP) method was applied to the three COG algorithms for the PET detector. Different powers of the RTP from 1.0 to 2.5 were evaluated. RESULTS: The proposed COG algorithm significantly improves the crystal identification at the junction of the two detectors as compared to the COG algorithm using four signals of each detector, and improves the crystal identification at the center of the two detectors as compared to the COG algorithm using eight signals from both detectors. The RTP method significantly improves the overall flood histogram qualities of the two COG algorithms using either eight or six signals from the two detectors, and the two COG algorithm provide similar flood histogram quality when a power of 1.5 is used. CONCLUSION: The novel positioning algorithms reduce the edge effect and improve the flood histogram quality for a PET detector consisting of two optically coupled detectors, each consisting of a pixelated scintillator crystal array and a SiPM array with highly multiplexed four signal readout. The positioning algorithms can be used in a PET scanner to improve the spatial resolution and sensitivity.

Enhancing pap smear image classification: integrating transfer learning and attention mechanisms for improved detection of cervical abnormalities.

Cervical cancer remains a major global health challenge, accounting for significant morbidity and mortality among women. Early detection through screening, such as Pap smear tests, is crucial for effective treatment and improved patient outcomes. However, traditional manual analysis of Pap smear images is labor-intensive, subject to human error, and requires extensive expertise. To address these challenges, automated approaches using deep learning techniques have been increasingly explored, offering the potential for enhanced diagnostic accuracy and efficiency. This research focuses on improving cervical cancer detection from Pap smear images using advanced deep-learning techniques. Specifically, we aim to enhance classification performance by leveraging Transfer Learning (TL) combined with an attention mechanism, supplemented by effective preprocessing techniques. Our preprocessing pipeline includes image normalization, resizing, and the application of Histogram of Oriented Gradients (HOG), all of which contribute to better feature extraction and improved model performance. The dataset used in this study is the Mendeley Liquid-Based Cytology (LBC) dataset, which provides a comprehensive collection of cervical cytology images annotated by expert cytopathologists. Initial experiments with the ResNet model on raw data yielded an accuracy of 63.95%. However, by applying our preprocessing techniques and integrating an attention mechanism, the accuracy of the ResNet model increased dramatically to 96.74%. Further, the Xception model, known for its superior feature extraction capabilities, achieved the best performance with an accuracy of 98.95%, along with high precision (0.97), recall (0.99), and F1-Score (0.98) on preprocessed data with an attention mechanism. These results underscore the effectiveness of combining preprocessing techniques, TL, and attention mechanisms to significantly enhance the performance of automated cervical cancer detection systems. Our findings demonstrate the potential of these advanced techniques to provide reliable, accurate, and efficient diagnostic tools, which could greatly benefit clinical practice and improve patient outcomes in cervical cancer screening.

Fractional order calculus model-derived histogram metrics for assessing pathological complete response to neoadjuvant chemotherapy in locally advanced rectal cancer.

AIM: This study evaluates the value of diffusion fractional order calculus (FROC) model for the assessment of pathological complete response (pCR) after neoadjuvant chemoradiotherapy (nCRT) for locally advanced rectal cancer (LARC) by using histogram analysis derived from whole-tumor volumes. MATERIALS AND METHODS: Ninety-eight patients were prospectively included. Every patient received MRI scans before and after nCRT using a 3.0-Tesla MRI machine. Parameters of the FROC model, including the anomalous diffusion coefficient (D), intravoxel diffusion heterogeneity (ß), spatial parameter (µ), and the standard apparent diffusion coefficient (ADC), were calculated. Changes in median values (ΔX-median) and ratio (rΔX-median) were calculated. Receiver operating characteristic (ROC) curves were used for evaluating the diagnostic performance. RESULTS: Pre-treatmentß-10th percentile values were significantly lower in the pCR group compared to the non-pCR group (p < 0.001). The Δß-median showed higher diagnostic accuracy (AUC = 0.870) and sensitivity (76.67 %) for predicting tumor response compared to MRI tumor regression grading (mrTRG) scores (AUC = 0.722; sensitivity = 90.0 %). DISCUSSION: The use of FROC alongside comprehensive tumor histogram analysis was found to be practical and effective in evaluating the tumor response to nCRT in LARC patients.

[Clinical Application of Equivalent Uniform Dose in Intensity-Modulated Rotational Radiotherapy Based on Eclipse TPS].

Objective: To investigate the application of equivalent uniform dose (EUD) in intensity-modulated rotational radiotherapy and to explore optimization methods for improving the quality of modulated treatment plans. Methods: The impact of the parameter a in the EUD formula on the characteristics of the EUD curve was analyzed using Python. Thirty cases of head and neck tumors, thoracic tumors, and pelvic tumors were randomly selected for treatment planning. Dose optimization for the target area and organs at risk were performed using a physics-based optimization approach or an optimization approach that combines physical constraints with the EUD function. The dose distribution and compliance with constraints of the two groups of plans were compared, while also observing the effect of different values of a on the planning outcomes. Results: The impact of the value of a on the changes in EUD curve characteristics was consistent with its impact on the results of EUD plan optimization. When -15≤ a≤-5, the dose distribution in the target area was more uniform; when 1≤ a≤7, the effect on the uniform dose and low-dose regions in organs at risk was more noticeable; when 10≤ a≤30, the effect of constraining the high-dose regions in organs at risk was more pronounced, with the EUD for the target area and organs at risk exhibiting different expressions under different a values. The study also found that the target dose distribution and the protection of organs at risk in the EUD optimization group were better than those in the physical optimization group only. Conclusion: The a-value has a significant impact on the, the dose distribution in the target area and the organ at risk, providing a reference for the setting of a-value while using EUD to optimize the intensity modulation plan. The using of EUD optimization method can not only achieve excellent dose distribution in the target area, but also significantly reduce the normal tissue dose and the probability of complications, which has certain clinical application value.

DeepCOVIDNet-CXR: deep learning strategies for identifying COVID-19 on enhanced chest X-rays.

OBJECTIVES: COVID-19 is one of the recent major epidemics, which accelerates its mortality and prevalence worldwide. Most literature on chest X-ray-based COVID-19 analysis has focused on multi-case classification (COVID-19, pneumonia, and normal) by the advantages of Deep Learning. However, the limited number of chest X-rays with COVID-19 is a prominent deficiency for clinical relevance. This study aims at evaluating COVID-19 identification performances using adaptive histogram equalization (AHE) to feed the ConvNet architectures with reliable lung anatomy of airways. METHODS: We experimented with balanced small- and large-scale COVID-19 databases using left lung, right lung, and complete chest X-rays with various AHE parameters. On multiple strategies, we applied transfer learning on four ConvNet architectures (MobileNet, DarkNet19, VGG16, and AlexNet). RESULTS: Whereas DarkNet19 reached the highest multi-case identification performance with an accuracy rate of 98.26 % on the small-scale dataset, VGG16 achieved the best generalization performance with an accuracy rate of 95.04 % on the large-scale dataset. CONCLUSIONS: Our study is one of the pioneering approaches that analyses 3615 COVID-19 cases and specifies the most responsible AHE parameters for ConvNet architectures in the multi-case classification.

Evaluating the ability of the Wind Erosion Prediction System (WEPS) to simulate near-surface wind speeds in the Inland Pacific Northwest, USA.

Wind speed is one of the main control factors of wind erosion and dust emissions, which are major problems in arid and semiarid regions of the world. Accurately simulating highly precise hourly wind speeds is critical and cost-efficient for land management decisions with the goal of mitigating wind erosion and land degradation. The Wind Erosion Prediction System (WEPS) is a process-based, daily time-step model that simulates changes in the soil-vegetation-atmosphere. However, to date, relatively few studies have been conducted to test the ability of the WEPS in simulating hourly wind speeds. In this study, the performance of the WEPS model was tested in the Inland Pacific Northwest (iPNW), where wind erosion is a serious problem. Hourly wind speeds were observed and simulated by the WEPS at 13 meteorological stations from 2009 to 2018 using the WEPS hourly wind speed probability histogram. Owing to increasing wind shear, the model is not as precise in reproducing high wind speeds. The WEPS inadequately simulated the hourly wind speeds at six of the 13 stations, with a low index of agreement (d < 0.5). The complex regional topography may be one of the reasons for this lack of agreement, because the WEPS's performance of interpolation relies on spatial distances and surface complexity. Therefore, we validated the model using another wind-speed database to eliminate the impact of spatial interpolation. The performance of the WEPS was improved after removing the impact of spatial interpolation, producing d values > 0.5 at nine of the 13 stations. Our results suggest that the WEPS can accurately simulate hourly wind speeds and assess wind erosion in the absence of interpolation, whereas the model may be uncertain when invoking spatial interpolation. Some evidence also suggests that the model may have a tendency to underestimate observed hourly wind speeds. Pragmatically, this suggests that model users should consider the possibility that WEPS may underestimate wind erosion risk in the iPNW and plan implementation of conservation practices accordingly.

Sinonasal adenoid cystic carcinoma: preoperative apparent diffusion coefficient histogram analysis in prediction of prognosis and Ki-67 proliferation status.

PURPOSE: To investigate the value of preoperative apparent diffusion coefficient (ADC) histogram analysis in predicting the prognosis of patients with sinonasal adenoid cystic carcinoma (ACC) and the correlation between ADC histogram parameters and Ki-67 labeling index (LI). MATERIALS AND METHODS: The study enrolled 66 patients with sinonasal ACC who were surgically resected and confirmed by histopathology. The disease-free survival (DFS) was evaluated with clinical-pathologic and radiologic characteristics using the Cox proportion hazard model. Spearman correlation analysis was used to evaluate the correlation between ADC histogram parameters and Ki-67 LI. The predictive performance of ADC histogram parameters for Ki-67 LI was assessed using the receiver operating characteristic (ROC) curve. RESULTS: Multivariable analysis showed Ki-67 LI (hazard ratio: 9.279; 95% confidence interval 1.099-78.338; P = 0.041) and ADCskewness (hazard ratio: 5.942; 95% confidence interval 1.832-19.268; P = 0.003) were significant independent predictors of DFS. The combination of these two variables achieved the predictive ability with a C-index of 0.717 (95% confidence interval 0.607-0.826). ADCmean and all ADC percentiles (10th, 50th, and 90th) significantly and inversely correlated with Ki-67 LI of ACC (Correlation coefficients = - 0.574 to - 0.591, Ps < 0.001). Among the ADC histogram parameters, the ADC50th showed superior performance for the differentiation of the high from low Ki-67 LI groups with an area under the curve (AUC) of 0.834 and an accuracy of 80.30%. CONCLUSION: ADC histogram analysis had predictive value for DFS and Ki-67 LI, which may be a valuable biomarker for prognosis and proliferation status for ACC in clinical practice.

Research on the Depth Image Reconstruction Algorithm Using the Two-Dimensional Kaniadakis Entropy Threshold.

The photon-counting light laser detection and ranging (LiDAR), especially the Geiger mode avalanche photon diode (Gm-APD) LiDAR, can obtain three-dimensional images of the scene, with the characteristics of single-photon sensitivity, but the background noise limits the imaging quality of the laser radar. In order to solve this problem, a depth image estimation method based on a two-dimensional (2D) Kaniadakis entropy thresholding method is proposed which transforms a weak signal extraction problem into a denoising problem for point cloud data. The characteristics of signal peak aggregation in the data and the spatio-temporal correlation features between target image elements in the point cloud-intensity data are exploited. Through adequate simulations and outdoor target-imaging experiments under different signal-to-background ratios (SBRs), the effectiveness of the method under low signal-to-background ratio conditions is demonstrated. When the SBR is 0.025, the proposed method reaches a target recovery rate of 91.7%, which is better than the existing typical methods, such as the Peak-picking method, Cross-Correlation method, and the sparse Poisson intensity reconstruction algorithm (SPIRAL), which achieve a target recovery rate of 15.7%, 7.0%, and 18.4%, respectively. Additionally, comparing with the SPIRAL, the reconstruction recovery ratio is improved by 73.3%. The proposed method greatly improves the integrity of the target under high-background-noise environments and finally provides a basis for feature extraction and target recognition.

Deep learning-based prediction of the dose-volume histograms for volumetric modulated arc therapy of left-sided breast cancer.

BACKGROUND: The advancements in artificial intelligence and computational power have made deep learning an attractive tool for radiotherapy treatment planning. Deep learning has the potential to significantly simplify the trial-and-error process involved in inverse planning required by modern treatment techniques such as volumetric modulated arc therapy (VMAT). In this study, we explore the ability of deep learning to predict organ-at-risk (OAR) dose-volume histograms (DVHs) of left-sided breast cancer patients undergoing VMAT treatment based solely on their anatomical characteristics. The predicted DVHs could be used to derive patient-specific dose constraints and dose objectives, streamlining the treatment planning process, standardizing the quality of the plans, and personalizing the treatment planning. PURPOSE: This study aimed to develop a deep learning-based framework for the prediction of organ-specific dose-volume histograms (DVH) based on structures delineated for left-sided breast cancer treatment. METHODS: We used a dataset of 249 left-sided breast cancer patients treated with tangential VMAT fields. We extracted delineated structures and dose distributions for each patient and derived slice-by-slice DVHs for planning target volume (PTV) and organs-at-risk. The patients were divided into training (70%, n = 174), validation (10%, n = 24), and test (20%, n = 51) sets. Collected data were used to train a deep learning model for the prediction of the DVHs based on the delineated structures. The developed deep learning model comprised a modified DenseNet architecture followed by a recurrent neural network. RESULTS: In the independent test set (n = 51), the point-wise differences in the slice-by-slice DVHs between the clinical and predicted DVHs were small; the mean squared errors were 3.53, 1.58, 2.28, 3.37, and 1.44 [×10-4] for PTV, heart, ipsilateral lung, contralateral lung, and contralateral breast, respectively. With the derived cumulative DVHs, the mean absolute difference ± standard deviation of mean doses between the clinical and the predicted DVH were 0.08 ± 0.04 Gy, 0.24 ± 0.22 Gy, 0.73 ± 0.46 Gy, 0.07 ± 0.06 Gy, and 0.14 ± 0.14 Gy for PTV, heart, ipsilateral lung, contralateral lung, and contralateral breast, respectively. CONCLUSIONS: The deep learning-based approach enabled automatic and reliable prediction of the DVH based on delineated structures. The predicted DVHs could potentially serve as patient-specific clinical goals used to aid treatment planning and avoid suboptimal plans or to derive optimization objectives and constraints for automated treatment planning.

Assessing the Histological Malignancy Grade of Olfactory Neuroblastoma Using the Apparent Diffusion Coefficient Histogram Analysis.

Introduction Olfactory neuroblastoma (ONB) is a rare malignant tumor of the upper nasal cavity. The Hyams classification is an important histological grading system for diagnosing recurrence and predicting survival in ONB. This study aimed to evaluate the utility of apparent diffusion coefficient (ADC) histogram analysis in distinguishing between high-grade and low-grade ONB based on the Hyams classification system. Methods This retrospective study included 17 patients (11 males, six females; mean age 54 years, range 29-84) diagnosed with ONB who underwent pretreatment magnetic resonance imaging (MRI) including diffusion-weighted imaging between December 2017 and September 2022. Two board-certified radiologists outlined the regions of interest on ADC maps of the tumors. Mean, minimum, maximum ADC, standard deviation, skewness, kurtosis, and entropy were calculated from the ADC histograms. Patients were divided into low-grade (Hyams I-II) and high-grade (Hyams III-IV) groups based on histopathological evaluation by a board-certified pathologist. ADC histogram parameters were compared between the two groups using Mann-Whitney U tests. Two-sided p-values of < 0.05 were considered statistically significant. Results The study included 10 low-grade (two grade I, eight grade II) and seven high-grade (five grade III, one grade III/IV, one grade IV) ONB cases. Comparison between the low-grade and high-grade groups showed no statistically significant differences in any of the ADC histogram parameters analyzed: mean ADC (median 1.02 vs 0.95; p = 0.591), minimum ADC (0.84 vs 0.78; p = 0.494), maximum ADC (1.06 vs 1.19; p = 0.625), standard deviation (0.09 vs 0.14; p = 0.433), skewness (-0.48 vs -0.75; p = 0.133), kurtosis (2.79 vs 3.12; p = 0.161), and entropy (4.69 vs 5.06; p = 0.315). Conclusion This study demonstrated that ADC histogram analysis was unable to differentiate between high-grade and low-grade ONB based on the Hyams classification. The findings suggest that preoperative grading of ONB malignancy using ADC histogram parameters is challenging. Thus, grading based on preoperative imaging evaluation is difficult.

Large-language-model empowered 3D dose prediction for intensity-modulated radiotherapy.

BACKGROUND: Treatment planning is currently a patient specific, time-consuming, and resource demanding task in radiotherapy. Dose-volume histogram (DVH) prediction plays a critical role in automating this process. The geometric relationship between DVHs in radiotherapy plans and organs-at-risk (OAR) and planning target volume (PTV) has been well established. This study explores the potential of deep learning models for predicting DVHs using images and subsequent human intervention facilitated by a large-language model (LLM) to enhance the planning quality. METHOD: We propose a pipeline to convert unstructured images to a structured graph consisting of image-patch nodes and dose nodes. A novel Dose Graph Neural Network (DoseGNN) model is developed for predicting DVHs from the structured graph. The proposed DoseGNN is enhanced with the LLM to encode massive knowledge from prescriptions and interactive instructions from clinicians. In this study, we introduced an online human-AI collaboration (OHAC) system as a practical implementation of the concept proposed for the automation of intensity-modulated radiotherapy (IMRT) planning. RESULTS: The proposed DoseGNN model was compared to widely employed DL models used in radiotherapy, including Swin Transformer, 3D U-Net CNN, and vanilla MLP. For PTV, DoseGNN achieved the mean absolute error (MAE) of D m a x ${D}_{max}$ , D m e a n ${D}_{mean}$ , D 95 ${D}_{95}$ , and D 1 ${D}_1$ between true plans and predicted plans that were 64%, 53%, 64%, 61% of the best baseline model. For the worst case among OARs (left lung, right lung, chest wall, heart, spinal cord), DoseGNN achieved the mean absolute error of D m a x ${D}_{max}$ , D m e a n ${D}_{mean}$ , D 50 ${D}_{50}$ that were 85%, 91%, 80% of the best baseline model. Moreover, the LLM-empowered DoseGNN model facilitates seamless adjustment to treatment plans through interaction with clinicians using natural language. CONCLUSION: We developed DoseGNN, a novel deep learning model for predicting delivered radiation doses from medical images, enhanced by LLM to allow adjustment through seamless interaction with clinicians. The preliminary results confirm DoseGNN's superior accuracy in DVH prediction relative to typical DL methods, highlighting its potential to facilitate an online clinician-AI collaboration system for streamlined treatment planning automation.

A high-precision genome size estimator based on the k-mer histogram correction.

Introduction: In the realm of next-generation sequencing datasets, various characteristics can be extracted through k-mer based analysis. Among these characteristics, genome size (GS) is one that can be estimated with relative ease, yet achieving satisfactory accuracy, especially in the context of heterozygosity, remains a challenge. Methods: In this study, we introduce a high-precision genome size estimator, GSET (Genome Size Estimation Tool), which is based on k-mer histogram correction. Results: We have evaluated GSET on both simulated and real datasets. The experimental results demonstrate that this tool can estimate genome size with greater precision, even surpassing the accuracy of state-of-the-art tools. Notably, GSET also performs satisfactorily on heterozygous datasets, where other tools struggle to produce useable results. Discussion: The processing model of GSET diverges from the popular data fitting models used by similar tools. Instead, it is derived from empirical data and incorporates a correction term to mitigate the impact of sequencing errors on genome size estimation. GSET is freely available for use and can be accessed at the following URL: https://github.com/Xingyu-Liao/GSET.

Application of a combined clinical prediction model based on enhanced T1-weighted image(T1WI) full volume histogram in peripheral nerve invasion (PNI) and lymphatic vessel invasion (LVI) in rectal cancer.

PURPOSE: This study aims to use a combined clinical prediction model based on enhanced T1-weighted image(T1WI) full volume histogram to predict preoperative peripheral nerve invasion (PNI) and lymphatic vessel invasion (LVI) in rectal cancer. METHODS: We included a total of 68 PNI patients and 80 LVI patients who underwent surgical resection and pathological confirmation of rectal cancer. According to the PNI/LVI status, patients were divided into PNI positive group (n = 39), the PNI negative group (n = 29), LVI positive group (n = 48), and the LVI negative group (n = 32). External validation included a total of 42 patients with nerve and vascular invasion in patients with surgically resected and pathologically confirmed rectal cancer at another healthcare facility, with a PNI positive group (n = 32) and a PNI-negative group (n = 10) as well as an LVI positive group (n = 35) and LVI-negative group (n = 7). All patients underwent 3.0T magnetic resonance T1WI enhanced scanning. We use Firevoxel software to delineate the region of interest (ROI), extract histogram parameters, and perform univariate analysis, LASSO regression, and multivariate logistic regression analysis in sequence to screen for the best predictive factors. Then, we constructed a clinical prediction model and plotted it into a column chart for personalized prediction. Finally, we evaluate the performance and clinical practicality of the model based on the area under curve (AUC), calibration curve, and decision curve. RESULTS: Multivariate logistic regression analysis found that variance and the 75th percentile were independent risk factors for PNI, while maximum and variance were independent risk factors for LVI. The clinical prediction model constructed based on the above factors has an AUC of 0.734 (95% CI: 0.591-0.878) for PNI in the training set and 0.731 (95% CI: 0.509-0.952) in the validation set; The training set AUC of LVI is 0.701 (95% CI: 0.561-0.841), and the validation set AUC is 0.685 (95% CI: 0.439-0.932). External validation showed an AUC of 0.722 (95% CI: 0.565-0.878) for PNI; and an AUC of 0.706 (95% CI: 0.481-0.931) for LVI. CONCLUSIONS: This study indicates that the combination of enhanced T1WI full volume histogram and clinical prediction model can be used to predict the perineural and lymphovascular invasion status of rectal cancer before surgery, providing valuable reference information for clinical diagnosis.

The diagnostic value of histogram analysis of DWI and DKI for the mismatch repair status of rectal adenocarcinoma.

Objectives: To compare the diagnostic value of histogram analysis derived from diffusion weighted imaging (DWI) and diffusion kurtosis imaging (DKI) in differentiating the mismatch repair (MMR) status of rectal adenocarcinoma. Methods: DWI and DKI were performed in 124 patients with rectal adenocarcinoma, which were divided into deficient mismatch repair (dMMR) group and proficient mismatch repair (pMMR) group. The patients' general clinical information, pathology and image characteristics were compared. The histogram analysis of apparent diffusion coefficient (ADC), diffusion kurtosis (K) and diffusion coefficient (D)derived from DWI and DKI at b values of 1000 and 2000 s/mm2 were calculated. The diagnostic efficacy of quantitative parameters for MMR in rectal adenocarcinoma was compared. Results: The mean, 50th, 75th and 90th in ADC quantitative parameters of dMMR group were lower when the b value was 2000 s/mm2 (all P < 0.05). With b value of 1000 s/mm2, the 10th, 25th, and 50th in the dMMR group were lower, and the skewness was higher (all P < 0.05). D values (10th, 25th and 50th) derived from DKI quantitative parameters were lower in the dMMR group. The K values (75th, 90th and Kskewness) were higher in the dMMR group, while Kkurtosis was lower (all P < 0.05). The results of multivariate logistic regression analysis showed that ADC75th(b = 2000 s/mm2), ADCskewness (b = 1000 s/mm2) and Kskewness were the statistical significant parameters (P = 0.014, 0.036 and 0.002, respectively), and the AUC values were 0.713, 0.818 and 0.835, respectively. Conclusion: Histogram analysis derived from DWI and DKI can be good predictor of MMR. Kskewness is the strongest independent factor for predicting MMR.

Effective removal of global tilt from atomically-resolved topography images of vicinal surfaces with narrow terraces.

The main feature of vicinal surfaces of crystals characterized by the Miller indices (hhm) is rather small width (less than 10 nm) and substantially large length (more than 200 nm) of atomically-flat terraces. This makes difficult to apply standard methods of image processing and correct visualization of crystalline lattices at the terraces and multiatomic steps. Here we consider two procedures allowing us to minimize effects of both small-scale noise and global tilt of sample: (i) analysis of the difference of two Gaussian blurred images, and (ii) subtraction of the plane, whose parameters are determined by optimization of the histogram of the visible heights, from raw topography image. It is shown that both methods provide nondistorted images demonstrating atomic structures on vicinal Si(556) and Si(557) surfaces.

Whole-lesion CT histogram analysis as an advanced technique in the portal venous phase: differentiating lipid poor adrenal adenomas from pheochromocytomas.

PURPOSE: Adrenal computed tomography (CT) has limitation due to imaging overlaps inthe washout characteristics of pheochromocytomas and adenomas (especially lipid-poor). The aim of this study was to investigate the distinguishability of lipid-poor adrenal adenomas and pheochromocytomas using whole-lesion CT histogram analysis. MATERIALS AND METHODS: Histopathologically proven 24 lipid-poor adenomas and 29 pheochromocytomas (total 53 lesions in 53 patients) were included in this retrospective study. Data obtained from standard and volumetric examinations of the lesions by dedicated adrenal CT were compared between the two groups using univariate analysis. Parameters that showed differences were further evaluated using multivariate logistic regression analysis. RESULTS: Univariate analysis revealed significant differences between the two groups in terms of lesion size, lesion volume, percentage of relative wash out, peak HU values and the percentage of voxels with attenuation ≥ 100 HU, ≥ 110 HU and ≥ 120 HU (p = 0.0001, P = 0.0001, P = 0.01, P = 0.008, p = 0.04, p = 0.02, p = 0.02, respectively). Multivariate analysis revealed lesion size ≥ 22.05 mm (OR: 22; p < 0.0001), the percentage of voxels with attenuation ≥ 120 HU being ≥ 9% (OR: 3.27; p = 0.04), peak HU value ≥ 161.5 HU (OR: 4.40; p = 0.01) as risk factors for pheochromocytomas. CONCLUSIONS: Whole lesion CT histogram analysis can be used to differentiate pheochromocytomas from lipid-poor adenomas. Lesion volume, the percentage of voxels with attenuation ≥ 120 HU and peak HU values are independent parameters that can assist in this differentiation. These findings may help avoid unnecessary biopsies and surgeries for lipid-poor adenomas, while identifying pheochromocytoma risk may improve perioperative patient management. Our results should be validated by future prospective studies.