Deep learning for colorectal cancer detection in contrast-enhanced CT without bowel preparation: a retrospective, multicentre study.

BACKGROUND: Contrast-enhanced CT scans provide a means to detect unsuspected colorectal cancer. However, colorectal cancers in contrast-enhanced CT without bowel preparation may elude detection by radiologists. We aimed to develop a deep learning (DL) model for accurate detection of colorectal cancer, and evaluate whether it could improve the detection performance of radiologists. METHODS: We developed a DL model using a manually annotated dataset (1196 cancer vs 1034 normal). The DL model was tested using an internal test set (98 vs 115), two external test sets (202 vs 265 in 1, and 252 vs 481 in 2), and a real-world test set (53 vs 1524). We compared the detection performance of the DL model with radiologists, and evaluated its capacity to enhance radiologists' detection performance. FINDINGS: In the four test sets, the DL model had the area under the receiver operating characteristic curves (AUCs) ranging between 0.957 and 0.994. In both the internal test set and external test set 1, the DL model yielded higher accuracy than that of radiologists (97.2% vs 86.0%, p < 0.0001; 94.9% vs 85.3%, p < 0.0001), and significantly improved the accuracy of radiologists (93.4% vs 86.0%, p < 0.0001; 93.6% vs 85.3%, p < 0.0001). In the real-world test set, the DL model delivered sensitivity comparable to that of radiologists who had been informed about clinical indications for most cancer cases (94.3% vs 96.2%, p > 0.99), and it detected 2 cases that had been missed by radiologists. INTERPRETATION: The developed DL model can accurately detect colorectal cancer and improve radiologists' detection performance, showing its potential as an effective computer-aided detection tool. FUNDING: This study was supported by National Science Fund for Distinguished Young Scholars of China (No. 81925023); Regional Innovation and Development Joint Fund of National Natural Science Foundation of China (No. U22A20345); National Natural Science Foundation of China (No. 82072090 and No. 82371954); Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application (No. 2022B1212010011); High-level Hospital Construction Project (No. DFJHBF202105).

FedDBL: Communication and Data Efficient Federated Deep-Broad Learning for Histopathological Tissue Classification.

Histopathological tissue classification is a fundamental task in computational pathology. Deep learning (DL)-based models have achieved superior performance but centralized training suffers from the privacy leakage problem. Federated learning (FL) can safeguard privacy by keeping training samples locally, while existing FL-based frameworks require a large number of well-annotated training samples and numerous rounds of communication which hinder their viability in real-world clinical scenarios. In this article, we propose a lightweight and universal FL framework, named federated deep-broad learning (FedDBL), to achieve superior classification performance with limited training samples and only one-round communication. By simply integrating a pretrained DL feature extractor, a fast and lightweight broad learning inference system with a classical federated aggregation approach, FedDBL can dramatically reduce data dependency and improve communication efficiency. Five-fold cross-validation demonstrates that FedDBL greatly outperforms the competitors with only one-round communication and limited training samples, while it even achieves comparable performance with the ones under multiple-round communications. Furthermore, due to the lightweight design and one-round communication, FedDBL reduces the communication burden from 4.6 GB to only 138.4 KB per client using the ResNet-50 backbone at 50-round training. Extensive experiments also show the scalability of FedDBL on model generalization to the unseen dataset, various client numbers, model personalization and other image modalities. Since no data or deep model sharing across different clients, the privacy issue is well-solved and the model security is guaranteed with no model inversion attack risk. Code is available at https://github.com/tianpeng-deng/FedDBL.

Artificial intelligence-based analysis of tumor-infiltrating lymphocyte spatial distribution for colorectal cancer prognosis.

BACKGROUND: Artificial intelligence (AI) technology represented by deep learning has made remarkable achievements in digital pathology, enhancing the accuracy and reliability of diagnosis and prognosis evaluation. The spatial distribution of CD3 + and CD8 + T cells within the tumor microenvironment has been demonstrated to have a significant impact on the prognosis of colorectal cancer (CRC). This study aimed to investigate CD3 CT (CD3 + T cells density in the core of the tumor [CT]) prognostic ability in patients with CRC by using AI technology. METHODS: The study involved the enrollment of 492 patients from two distinct medical centers, with 358 patients assigned to the training cohort and an additional 134 patients allocated to the validation cohort. To facilitate tissue segmentation and T-cells quantification in whole-slide images (WSIs), a fully automated workflow based on deep learning was devised. Upon the completion of tissue segmentation and subsequent cell segmentation, a comprehensive analysis was conducted. RESULTS: The evaluation of various positive T cell densities revealed comparable discriminatory ability between CD3 CT and CD3-CD8 (the combination of CD3 + and CD8 + T cells density within the CT and invasive margin) in predicting mortality (C-index in training cohort: 0.65 vs. 0.64; validation cohort: 0.69 vs. 0.69). The CD3 CT was confirmed as an independent prognostic factor, with high CD3 CT density associated with increased overall survival (OS) in the training cohort (hazard ratio [HR] = 0.22, 95% confidence interval [CI]: 0.12-0.38, P <0.001) and validation cohort (HR = 0.21, 95% CI: 0.05-0.92, P = 0.037). CONCLUSIONS: We quantify the spatial distribution of CD3 + and CD8 + T cells within tissue regions in WSIs using AI technology. The CD3 CT confirmed as a stage-independent predictor for OS in CRC patients. Moreover, CD3 CT shows promise in simplifying the CD3-CD8 system and facilitating its practical application in clinical settings.

Alginate-CaCO3 hybrid colloidal hydrogel with tunable physicochemical properties for cell growth.

Biomaterials composed of food polysaccharides are of great interest for future biomedical applications due to their great biocompatibility, tunable mechanical properties, and complex architectural designs that play a crucial role in the modulation of cell adhesion and proliferation. In this work, a facile approach was designed to obtain novel 3D alginate-CaCO3 hybrid hydrogel particles in situ. Controlling the gel concentration from 3 to 20 mg·mL-1 allows us to control the alginate-CaCO3 hydrogel particles' size and density (size variation from 1.86 to 2.34 mm and density from 1.22 to 1.29 mg/mm3). This variable also has a considerable influence on the mineralization process resulting in CaCO3 particles with varied sizes and amounts within the hydrogel beads. The measurements of Young's modulus showed that the inclusion of CaCO3 particles into the alginate hydrogel improved its mechanical properties, and Young's modulus of these hybrid hydrogel particles had a linear relationship with alginate content and hydrogel particle size. Cell experiments indicated that alginate-CaCO3 hybrid hydrogel particles can support osteoblastic cell proliferation and growth. In particular, the amount of hydroxyapatite deposition on the cell membrane significantly increased after the treatment of cells with hybrid hydrogel particles, up to 20-fold. This work offers a strategy for constructing inorganic particle-doped polysaccharide hybrid hydrogel scaffolds that provide the potential to support cell growth.

Non-invasive prediction of microsatellite instability in colorectal cancer by a genetic algorithm-enhanced artificial neural network-based CT radiomics signature.

OBJECTIVE: The stratification of microsatellite instability (MSI) status assists clinicians in making treatment decisions for colorectal cancer (CRC) patients. This study aimed to establish a CT-based radiomics signature to predict MSI status in patients with CRC. METHODS: A total of 837 CRC patients who underwent preoperative enhanced CT and had available MSI status data were recruited from two hospitals. Radiomics features were extracted from segmented tumours, and a series of data balancing and feature selection strategies were used to select MSI-related features. Finally, an MSI-related radiomics signature was constructed using a genetic algorithm-enhanced artificial neural network model. Combined and clinical models were constructed using multivariate logistic regression analyses by integrating the clinical factors with or without the signature. A Kaplan-Meier survival analysis was conducted to explore the prognostic information of the signature in patients with CRC. RESULTS: Ten features were selected to construct a signature which showed robust performance in both the internal and external validation cohorts, with areas under the curves (AUC) of 0.788 and 0.775, respectively. The performance of the signature was comparable to that of the combined model (AUCs of 0.777 and 0.767, respectively) and it outperformed the clinical model constituting age and tumour location (AUCs of 0.768 and 0.623, respectively). Survival analysis demonstrated that the signature could stratify patients with stage II CRC according to prognosis (HR: 0.402, p = 0.029). CONCLUSIONS: This study built a robust radiomics signature for identifying the MSI status of CRC patients, which may assist individualised treatment decisions. KEY POINTS: • Our well-designed modelling strategies helped overcome the problem of data imbalance caused by the low incidence of MSI. • Genetic algorithm-enhanced artificial neural network-based CT radiomics signature can effectively distinguish the MSI status of CRC patients. • Kaplan-Meier survival analysis demonstrated that our signature could significantly stratify stage II CRC patients into high- and low-risk groups.

Effect of rapid rehabilitation nursing on inflammation and liver function after laparoscopic radical resection of primary liver cancer.

OBJECTIVE: To explore the effect of rapid rehabilitation nursing on inflammation and liver function in patients with primary liver cancer (PLC) after laparoscopic radical resection. METHODS: A total of 124 PLC patients who underwent laparoscopic radical surgery in the Zhuji People's Hospital of Zhejiang Province from April 2019 to July 2021 were enrolled in this retrospective study. Among them, 65 patients who received rapid rehabilitation nursing were assigned into the observation group (OG), and the other 59 with routine nursing were considered to be the control group (CG). The pain before operation (T0), 3 days after operation (T1) and 7 days after operation (T2) was evaluated by visual analogue scale (VAS). The perioperative related indexes and nursing satisfaction were compared. The levels of liver function indexes alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL) and inflammatory factors C-reactive protein (CRP), interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) were measured before operation, 1 day and 7 days after operation. Finally, the incidence of postoperative complications was counted, the 6-month survival rate of both groups of patients was compared. RESULTS: There was no obvious difference in VAS scores between the two groups at T0 (P>0.050), but the VAS scores at T1 and T2 in the OG were lower than those in the CG (P<0.001). There was no marked difference in the total operation time. Compared with the CG, the time to first exhaust, catheter indwelling and hospitalization in the OG were shorter (P<0.001) and the nursing satisfaction rate was higher (P<0.05). There was no obvious difference in ALT, AST and TBIL on the 1st day after operation (P>0.05); however, on the 7th day after operation, ALT and AST were lower while TBIL was higher in the OG (all P<0.05). There was no marked difference in CRP, IL-6 and TNF-α between the two groups on postoperative day 1 (P>0.05), but the levels were lower in the OG than those in the CG on postoperative day 7 (all P<0.05), and the total incidence of adverse reactions in the OG was lower (P<0.05). There was no statistical difference in the postoperative survival rate between both groups of patients (P>0.05). Age, number of lesions, tumor size, Child-Pugh grade, AST, TBIL, CRP, IL-6, TNF-α were associated with the survival rate of patients. CONCLUSION: Rapid rehabilitation nursing can effectively reduce adverse reactions after laparoscopic radical resection of PLC. Thus, it has a high application value in future clinical treatment.

A New Deep Learning Method with Self-Supervised Learning for Delineation of the Electrocardiogram.

Heartbeat characteristic points are the main features of an electrocardiogram (ECG), which can provide important information for ECG-based cardiac diagnosis. In this manuscript, we propose a self-supervised deep learning framework with modified Densenet to detect ECG characteristic points, including the onset, peak and termination points of P-wave, QRS complex wave and T-wave. We extracted high-level features of ECG heartbeats from the QT Database (QTDB) and two other larger datasets, MIT-BIH Arrhythmia Database (MITDB) and MIT-BIH Normal Sinus Rhythm Database (NSRDB) with no human-annotated labels as pre-training. By applying different transformations to ECG signals, the task of discriminating signals before and after transformation was defined as the pretext task. Subsequently, the convolutional layer was frozen and the weights of the self-supervised network were transferred to the downstream task of characteristic point localizations on heart beats in the QT dataset. Finally, the mean ± standard deviation of the detection errors of our proposed self-supervised learning method in QTDB for detecting the onset, peak, and termination points of P-waves, the onset and termination points of QRS waves, and the peak and termination points of T-waves were -0.24 ± 10.04, -0.48 ± 11.69, -0.28 ± 10.19, -3.72 ± 8.18, -4.12 ± 13.54, -0.68 ± 20.42, and 1.34 ± 21.04. The results show that the deep learning network based on the self-supervised framework constructed in this manuscript can accurately detect the feature points of a heartbeat, laying the foundation for automatic extraction of key information related to ECG-based diagnosis.

Nanoarchitectonics beyond perfect order - not quite perfect but quite useful.

Nanoarchitectonics, like architectonics, allows the design and building of structures, but at the nanoscale. Unlike those in architectonics, and even macro-, micro-, and atomic-scale architectonics, the assembled structures at the nanoscale do not always follow the projected design. In fact, they do follow the projected design but only for self-assembly processes producing structures with perfect order. Here, we look at nanoarchitectonics allowing the building of nanostructures without a perfect arrangement of building blocks. Here, fabrication of structures from molecules, polymers, nanoparticles, and nanosheets to polymer brushes, layer-by-layer assembly structures, and hydrogels through self-assembly processes is discussed, where perfect order is not necessarily the aim to be achieved. Both planar substrate and spherical template-based assemblies are discussed, showing the challenging nature of research in this field and the usefulness of such structures for numerous applications, which are also discussed here.

Hard, Soft, and Hard-and-Soft Drug Delivery Carriers Based on CaCO3 and Alginate Biomaterials: Synthesis, Properties, Pharmaceutical Applications.

Because free therapeutic drug molecules often have adverse effects on normal tissues, deliver scanty drug concentrations and exhibit a potentially low efficacy at pathological sites, various drug carriers have been developed for preclinical and clinical trials. Their physicochemical and toxicological properties are the subject of extensive research. Inorganic calcium carbonate particles are promising candidates as drug delivery carriers owning to their hardness, porous internal structure, high surface area, distinctive pH-sensitivity, low degradability, etc, while soft organic alginate hydrogels are also widely used because of their special advantages such as a high hydration, bio-adhesiveness, and non-antigenicity. Here, we review these two distinct substances as well as hybrid structures encompassing both types of carriers. Methods of their synthesis, fundamental properties and mechanisms of formation, and their respective applications are described. Furthermore, we summarize and compare similarities versus differences taking into account unique advantages and disadvantages of these drug delivery carriers. Moreover, rational combination of both carrier types due to their performance complementarity (yin-&yang properties: in general, yin is referred to for definiteness as hard, and yang is broadly taken as soft) is proposed to be used in the so-called hybrid carriers endowing them with even more advanced properties envisioned to be attractive for designing new drug delivery systems.

Development and validation of a computed tomography-based immune ecosystem diversity index as an imaging biomarker in non-small cell lung cancer.

OBJECTIVES: To date, there are no data on the noninvasive surrogate of intratumoural immune status that could be prognostic of survival outcomes in non-small cell lung cancer (NSCLC). We aimed to develop and validate the immune ecosystem diversity index (iEDI), an imaging biomarker, to indicate the intratumoural immune status in NSCLC. We further investigated the clinical relevance of the biomarker for survival prediction. METHODS: In this retrospective study, two independent NSCLC cohorts (Resec1, n = 149; Resec2, n = 97) were included to develop and validate the iEDI to classify the intratumoural immune status. Paraffin-embedded resected specimens in Resec1 and Resec2 were stained by immunohistochemistry, and the density percentiles of CD3+, CD4+, and CD8+ T cells to all cells were quantified to estimate intratumoural immune status. Then, EDI features were extracted using preoperative computed tomography to develop an imaging biomarker, called iEDI, to determine the immune status. The prognostic value of iEDI was investigated on NSCLC patients receiving surgical resection (Resec1; Resec2; internal cohort Resec3, n = 419; external cohort Resec4, n = 96; and TCIA cohort Resec5, n = 55). RESULTS: iEDI successfully classified immune status in Resec1 (AUC 0.771, 95% confidence interval [CI] 0.759-0.783; and 0.770 through internal validation) and Resec2 (0.669, 0.647-0.691). Patients with higher iEDI-score had longer overall survival (OS) in Resec3 (unadjusted hazard ratio 0.335, 95%CI 0.206-0.546, p < 0.001), Resec4 (0.199, 0.040-1.000, p < 0.001), and TCIA (0.303, 0.098-0.944, p = 0.001). CONCLUSIONS: iEDI is a non-invasive surrogate of intratumoural immune status and prognostic of OS for NSCLC patients receiving surgical resection. KEY POINTS: • Decoding tumour immune microenvironment enables advanced biomarkers identification. • Immune ecosystem diversity index characterises intratumoural immune status noninvasively. • Immune ecosystem diversity index is prognostic for NSCLC patients.

Improving small animal cone beam CT resolution by mitigating x-ray focal spot induced blurring via deconvolution.

Objective.Modern preclinical small animal radiation platforms utilize cone beam computerized tomography (CBCT) for image guidance and experiment planning purposes. The resolution of CBCT images is of particular importance for visualizing fine animal anatomical structures. One major cause of spatial resolution reduction is the finite size of the x-ray focal spot. In this work, we proposed a simple method to measure x-ray focal spot intensity map and a CBCT image domain deblurring model to mitigate the effect of focal spot-induced image blurring.Approach.We measured a projection image of a tungsten ball bearing using the flat panel detector of the CBCT platform. We built a forward blurring model of the projection image and derived the spot intensity map by deconvolving the measured projection image. Based on the measured spot intensity map, we derived a CBCT image domain blurring model for images reconstructed by the filtered backprojection algorithm. Based on this model, we computed image domain blurring kernel and improved the CBCT image resolution by deconvolving the CBCT image.Main results.We successfully measured the x-ray focal spot intensity map. The spot size characterized by full width at half maximum was ∼0.75 × 0.55 mm2at 40 kVp. We computed image domain convolution kernels caused by the x-ray focal spot. A simulation study on noiseless projections was performed to evaluate the spatial resolution improvement exclusively by the focal spot kernel, and the modulation transfer function (MTF) at 50% was increased from 1.40 to 1.65 mm-1for in-plane images and 1.05-1.32 mm-1for cross-plane images. Experimental studies on a CT insert phantom and a plastinated mouse phantom demonstrated improved spatial resolution after image domain deconvolution, as indicated by visually improved resolution of fine structures. MTF at 50% was improved from 1.00 to 1.12 mm-1for in-plane direction and from 0.72 to 0.84 mm-1for cross-plane direction.Significance.The proposed method to mitigate blurring caused by finite x-ray spot size and improve CBCT image resolution is simple and effective.

Development and Validation of an Immune-Based Prognostic Risk Score for Patients With Resected Non-Small Cell Lung Cancer.

Background: Despite the well-known role of immunoscore, as a prognostic tool, that appeared to be superior to tumor-node-metastasis (TNM) staging system, no prognostic scoring system based on immunohistochemistry (IHC) staining digital image analysis has been established in non-small cell lung cancer (NSCLC). Hence, we aimed to develop and validate an immune-based prognostic risk score (IMPRS) that could markedly improve individualized prediction of postsurgical survival in patients with resected NSCLC. Methods: In this retrospective study, complete resection of NSCLC (stage I-IIIA) was performed for two independent patient cohorts (discovery cohort, n=168; validation cohort, n=115). Initially, paraffin-embedded resected specimens were stained by immunohistochemistry (IHC) of three immune cell types (CD3+, CD4+, and CD8+ T cells), and a total of 5,580 IHC-immune features were extracted from IHC digital images for each patient by using fully automated pipeline. Then, an IHC-immune signature was constructed with selected features using the LASSO Cox analysis, and the association of signature with patients' overall survival (OS) was analyzed by Kaplan-Meier method. Finally, IMPRS was established by incorporating IHC-immune signature and independent clinicopathological variables in multivariable Cox regression analysis. Furthermore, an external validation cohort was included to validate this prognostic risk score. Results: Eight key IHC-immune features were selected for the construction of IHC-immune signature, which showed significant associations with OS in all cohorts [discovery: hazard ratio (HR)=11.518, 95%CI, 5.444-24.368; validation: HR=2.664, 95%CI, 1.029-6.896]. Multivariate analyses revealed IHC-immune signature as an independent prognostic factor, and age, T stage, and N stage were also identified and entered into IMPRS (all p<0.001). IMPRS had good discrimination ability for predicting OS (C-index, 0.869; 95%CI, 0.861-0.877), confirmed using external validation cohort (0.731, 0.717-0.745). Interestingly, IMPRS had better prognostic value than clinicopathological-based model and TNM staging system termed as C-index (clinicopathological-based model: 0.674; TNM staging: 0.646, all p<0.05). More importantly, decision curve analysis showed that IMPRS had adequate performance for predicting OS in resected NSCLC patients. Conclusions: Our findings indicate that the IMPRS that we constructed can provide more accurate prognosis for individual prediction of OS for patients with resected NSCLC, which can help in guiding personalized therapy and improving outcomes for patients.

Coupling radiomics analysis of CT image with diversification of tumor ecosystem: A new insight to overall survival in stage I-III colorectal cancer.

Objective: This study aimed to establish a method to predict the overall survival (OS) of patients with stage I-III colorectal cancer (CRC) through coupling radiomics analysis of CT images with the measurement of tumor ecosystem diversification. Methods: We retrospectively identified 161 consecutive patients with stage I-III CRC who had underwent radical resection as a training cohort. A total of 248 patients were recruited for temporary independent validation as external validation cohort 1, with 103 patients from an external institute as the external validation cohort 2. CT image features to describe tumor spatial heterogeneity leveraging the measurement of diversification of tumor ecosystem, were extracted to build a marker, termed the EcoRad signature. Multivariate Cox regression was used to assess the EcoRad signature, with a prediction model constructed to demonstrate its incremental value to the traditional staging system for OS prediction. Results: The EcoRad signature was significantly associated with OS in the training cohort [hazard ratio (HR)=6.670; 95% confidence interval (95% CI): 3.433-12.956; P<0.001), external validation cohort 1 (HR=2.866; 95% CI: 1.646-4.990; P<0.001) and external validation cohort 2 (HR=3.342; 95% CI: 1.289-8.663; P=0.002). Incorporating the EcoRad signature into the prediction model presented a higher prediction ability (P<0.001) with respect to the C-index (0.813, 95% CI: 0.804-0.822 in the training cohort; 0.758, 95% CI: 0.751-0.765 in the external validation cohort 1; and 0.746, 95% CI: 0.722-0.770 in external validation cohort 2), compared with the reference model that only incorporated tumor, node, metastasis (TNM) system, as well as a better calibration, improved reclassification and superior clinical usefulness. Conclusions: This study establishes a method to measure the spatial heterogeneity of CRC through coupling radiomics analysis with measurement of diversification of the tumor ecosystem, and suggests that this approach could effectively predict OS and could be used as a supplement for risk stratification among stage I-III CRC patients.

Rate dependence of early afterdepolarizations in the His-Purkinje system: A simulation study.

BACKGROUND AND OBJECTIVE: Early afterdepolarizations (EADs) are associated with a variety of arrhythmias and have the property of rate dependence. EADs can occur in Purkinje cells while the effect of rate dependence of EADs in the His-Purkinje system has not been fully investigated. In order to reveal the rate dependence of EADs in the His-Purkinje system and its effect on ventricular electrical activities, the simulation research was carried out in this manuscript. METHODS: This manuscript first studied the relationship between the occurrence of EADs and stimulation cycle length on the DiFranNoble cell model. Then, the relationship between the rate dependence of EADs and the conduction block of the His-Purkinje system at slow heart rates was studied on the rabbit whole ventricular model including the His-Purkinje system, and its mechanism was analyzed from multiple angles. RESULTS: ① The rate dependence of EADs is related to the inconsistency of EADs occurrence in the His-Purkinje system. When the stimulation cycle length is long or short enough, EADs either occur or not occur stably in the His-Purkinje system, while in a certain stimulation cycle length window, the chaotic state of EADs will be observed. ② The key subcellular factors x-gate is an important mechanism involved to the rate dependence of EADs in the His-Purkinje system. ③ The discrete distribution of x-gate values and the "source-sink" mechanism lead to the inconsistency of EADs in the His-Purkinje system. The prolonged action potential duration caused by EADs can lead to conduction block at slow heart rates. CONCLUSION: The rate dependence of EADs in Purkinje system can lead to disordered ventricular electrical activity.

Improving dose calculation accuracy in preclinical radiation experiments using multi-energy element resolved cone-beam CT.

Objective.Cone-beam CT (CBCT) in modern pre-clinical small-animal radiation research platforms provides volumetric images for image guidance and experiment planning purposes. In this work, we implemented multi-energy element-resolved (MEER) CBCT using three scans with different kVps on a SmART platform (Precision x-ray Inc.) to determine images of relative electron density (rED) and elemental composition (EC) that are needed for Monte Carlo-based radiation dose calculation.Approach.We performed comprehensive calibration tasks to achieve sufficient accuracy for this quantitative imaging purpose. For geometry calibration, we scanned a ball bearing phantom and used an analytical method together with an optimization approach to derive gantry angle specific geometry parameters. Intensity calibration and correction included the corrections for detector lag, glare, and beam hardening. The corrected CBCT projection images acquired at 30, 40, and 60 kVp in multiple scans were used to reconstruct CBCT images using the Feldkamp-Davis-Kress reconstruction algorithm. After that, an optimization problem was solved to determine images of rED and EC. We demonstrated the effectiveness of our CBCT calibration steps by showing improvements in image quality and successful material decomposition in cases with a small animal CT calibration phantom and a plastinated mouse phantom.Main results.It was found that artifacts induced by geometry inaccuracy, detector lag, glare, and beam hardening were visually reduced. CT number mean errors were reduced from 19% to 5%. In the CT calibration phantom case, median errors in H, O, and Ca fractions for all the inserts were below 1%, 2%, and 4% respectively, and median error in rED was less than 5%. Compared to the standard approach deriving material type and rED via CT number conversion, our approach improved Monte Carlo simulation-based dose calculation accuracy in bone regions. Mean dose error was reduced from 47.5% to 10.9%.Significance.The MEER-CBCT implemented on an existing CBCT system of a small animal irradiation platform achieved accurate material decomposition and significantly improved Monte Carlo dose calculation accuracy.

A radiomics prognostic scoring system for predicting progression-free survival in patients with stage IV non-small cell lung cancer treated with platinum-based chemotherapy.

OBJECTIVE: To develop and validate a radiomics prognostic scoring system (RPSS) for prediction of progression-free survival (PFS) in patients with stage IV non-small cell lung cancer (NSCLC) treated with platinum-based chemotherapy. METHODS: In this retrospective study, four independent cohorts of stage IV NSCLC patients treated with platinum-based chemotherapy were included for model construction and validation (Discovery: n=159; Internal validation: n=156; External validation: n=81, Mutation validation: n=64). First, a total of 1,182 three-dimensional radiomics features were extracted from pre-treatment computed tomography (CT) images of each patient. Then, a radiomics signature was constructed using the least absolute shrinkage and selection operator method (LASSO) penalized Cox regression analysis. Finally, an individualized prognostic scoring system incorporating radiomics signature and clinicopathologic risk factors was proposed for PFS prediction. RESULTS: The established radiomics signature consisting of 16 features showed good discrimination for classifying patients with high-risk and low-risk progression to chemotherapy in all cohorts (All P<0.05). On the multivariable analysis, independent factors for PFS were radiomics signature, performance status (PS), and N stage, which were all selected into construction of RPSS. The RPSS showed significant prognostic performance for predicting PFS in discovery [C-index: 0.772, 95% confidence interval (95% CI): 0.765-0.779], internal validation (C-index: 0.738, 95% CI: 0.730-0.746), external validation (C-index: 0.750, 95% CI: 0.734-0.765), and mutation validation (C-index: 0.739, 95% CI: 0.720-0.758). Decision curve analysis revealed that RPSS significantly outperformed the clinicopathologic-based model in terms of clinical usefulness (All P<0.05). CONCLUSIONS: This study established a radiomics prognostic scoring system as RPSS that can be conveniently used to achieve individualized prediction of PFS probability for stage IV NSCLC patients treated with platinum-based chemotherapy, which holds promise for guiding personalized pre-therapy of stage IV NSCLC.

A deep learning quantified stroma-immune score to predict survival of patients with stage II-III colorectal cancer.

BACKGROUND: Profound heterogeneity in prognosis has been observed in colorectal cancer (CRC) patients with intermediate levels of disease (stage II-III), advocating the identification of valuable biomarkers that could improve the prognostic stratification. This study aims to develop a deep learning-based pipeline for fully automatic quantification of immune infiltration within the stroma region on immunohistochemical (IHC) whole-slide images (WSIs) and further analyze its prognostic value in CRC. METHODS: Patients from two independent cohorts were divided into three groups: the development group (N = 200), the internal (N = 134), and the external validation group (N = 90). We trained a convolutional neural network for tissue classification of CD3 and CD8 stained WSIs. A scoring system, named stroma-immune score, was established by quantifying the density of CD3+ and CD8+ T-cells infiltration in the stroma region. RESULTS: Patients with higher stroma-immune scores had much longer survival. In the development group, 5-year survival rates of the low and high scores were 55.7% and 80.8% (hazard ratio [HR] for high vs. low 0.39, 95% confidence interval [CI] 0.24-0.63, P < 0.001). These results were confirmed in the internal and external validation groups with 5-year survival rates of low and high scores were 57.1% and 78.8%, 63.9% and 88.9%, respectively (internal: HR for high vs. low 0.49, 95% CI 0.28-0.88, P = 0.017; external: HR for high vs. low 0.35, 95% CI 0.15-0.83, P = 0.018). The combination of stroma-immune score and tumor-node-metastasis (TNM) stage showed better discrimination ability for survival prediction than using the TNM stage alone. CONCLUSIONS: We proposed a stroma-immune score via a deep learning-based pipeline to quantify CD3+ and CD8+ T-cells densities within the stroma region on WSIs of CRC and further predict survival.

Prognostic value of a modified Immunoscore in patients with stage I-III resectable colon cancer.

OBJECTIVE: The Immunoscore method has proved fruitful for predicting prognosis in patients with colon cancer. However, there is still room for improvement in this scoring method to achieve further advances in its clinical translation. This study aimed to develop and validate a modified Immunoscore (IS-mod) system for predicting overall survival (OS) in patients with stage I-III colon cancer. METHODS: The IS-mod was proposed by counting CD3+ and CD8+ immune cells in regions of the tumor core and its invasive margin by drawing two lines of interest. A discovery cohort (N=212) and validation cohort (N=103) from two centers were used to evaluate the prognostic value of the IS-mod. RESULTS: In the discovery cohort, 5-year survival rates were 88.6% in the high IS-mod group and 60.7% in the low IS-mod group. Multivariate analysis confirmed that the IS-mod was an independent prognostic factor for OS [adjusted hazard ratio (HR)=0.36, 95% confidence interval (95% CI): 0.20-0.63]. With less annotation and computation cost, the IS-mod achieved performance comparable to that of the Immunoscore-like (IS-like) system (C-index, 0.676 vs. 0.661, P=0.231). The 2-category IS-mod using 47.5% as the threshold had a better prognostic value than that using a fixed threshold of 25% (C-index, 0.653 vs. 0.573, P=0.004). Similar results were confirmed in the validation cohort. CONCLUSIONS: Our method simplifies the annotation and accelerates the calculation of Immunoscore method, thus making it easier for clinical implementation. The IS-mod achieved comparable prognostic performance when compared to the IS-like system in both cohorts. Besides, we further found that even with a small reference set (N≥120), the IS-mod still demonstrated a stable prognostic value. This finding may inspire other institutions to develop a local reference set of an IS-mod system for more accurate risk stratification of colon cancer.

The value of the tumour-stroma ratio for predicting neoadjuvant chemoradiotherapy response in locally advanced rectal cancer: a case control study.

BACKGROUND: The tumour-stroma ratio (TSR) is recognized as a practical prognostic factor in colorectal cancer. However, TSR assessment generally utilizes surgical specimens. This study aims to investigate whether the TSR evaluated from preoperative biopsy specimens by a semi-automatic quantification method can predict the response after neoadjuvant chemoradiotherapy (nCRT) of patients with locally advanced rectal cancer (LARC). METHODS: A total of 248 consecutive patients diagnosed with LARC and treated with nCRT followed by resection were included. Haematoxylin and eosin (HE)-stained sections of biopsy specimens were collected, and the TSR was evaluated by a semi-automatic quantification method and was divided into three categories, using the cut-offs determined in the whole cohort to balance the proportion of patients in each category. The response to nCRT was evaluated on the primary tumour resection specimen by an expert pathologist using the four-tier tumour regression grade (TRG) system. RESULTS: The TSR can discriminate patients that are major-responders (TRG 0-1) from patients that are non-responders (TRG 2-3). Patients were divided into stroma-low (33.5%), stroma-intermediate (33.9%), and stroma-high (32.7%) groups using 56.3 and 72.8% as the cutoffs. In the stroma-low group, 58 (69.9%) patients were major-responders, and only 39 (48.1%) patients were considered major-responders in the stroma-high group (P = 0.018). Multivariate analysis showed that the TSR was the only pre-treatment predictor of response to nCRT (adjusted odds ratio 0.40, 95% confidence interval 0.21-0.76, P = 0.002). CONCLUSION: An elevated TSR in preoperative biopsy specimens is an independent predictor of nCRT response in LARC. This semi-automatic quantified TSR could be easily translated into routine pathologic assessment due to its reproducibility and reliability.