Quantifying tumour heterogeneity in 18F-FDG PET/CT imaging by texture analysis.

(18)F-Fluorodeoxyglucose positron emission tomography/computed tomography ((18)F-FDG PET/CT) is now routinely used in oncological imaging for diagnosis and staging and increasingly to determine early response to treatment, often employing semiquantitative measures of lesion activity such as the standardized uptake value (SUV). However, the ability to predict the behaviour of a tumour in terms of future therapy response or prognosis using SUVs from a baseline scan prior to treatment is limited. It is recognized that medical images contain more useful information than may be perceived with the naked eye, leading to the field of "radiomics" whereby additional features can be extracted by computational postprocessing techniques. In recent years, evidence has slowly accumulated showing that parameters obtained by texture analysis of radiological images, reflecting the underlying spatial variation and heterogeneity of voxel intensities within a tumour, may yield additional predictive and prognostic information. It is hoped that measurement of these textural features may allow better tissue characterization as well as better stratification of treatment in clinical trials, or individualization of future cancer treatment in the clinic, than is possible with current imaging biomarkers. In this review we focus on the literature describing the emerging methods of texture analysis in (18)FDG PET/CT, as well as other imaging modalities, and how the measurement of spatial variation of voxel grey-scale intensity within an image may provide additional predictive and prognostic information, and postulate the underlying biological mechanisms.

An efficient and robust Phonocardiography (PCG)-based Valvular Heart Diseases (VHD) detection framework using Vision Transformer (ViT).

BACKGROUND AND OBJECTIVES: Valvular heart diseases (VHDs) are one of the dominant causes of cardiovascular abnormalities that have been associated with high mortality rates globally. Rapid and accurate diagnosis of the early stage of VHD based on cardiac phonocardiogram (PCG) signal is critical that allows for optimum medication and reduction of mortality rate. METHODS: To this end, the current study proposes novel deep learning (DL)-based high-performance VHD detection frameworks that are relatively simpler in terms of network structures, yet effective for accurately detecting multiple VHDs. We present three different frameworks considering both 1D and 2D PCG raw signals. For 1D PCG, Mel frequency cepstral coefficients (MFCC) and linear prediction cepstral coefficients (LPCC) features, whereas, for 2D PCG, various deep convolutional neural networks (D-CNNs) features are extracted. Additionally, nature/bio-inspired algorithms (NIA/BIA) including particle swarm optimization (PSO) and genetic algorithm (GA) have been utilized for automatic and efficient feature selection directly from the raw PCG signal. To further improve the performance of the classifier, vision transformer (ViT) has been implemented levering the self-attention mechanism on the time frequency representation (TFR) of 2D PCG signal. Our extensive study presents a comparative performance analysis and the scope of enhancement for the combination of different descriptors, classifiers, and feature selection algorithms. MAIN RESULTS: Among all classifiers, ViT provides the best performance by achieving mean average accuracy Acc of 99.90 % and F1-score of 99.95 % outperforming current state-of-the-art VHD classification models. CONCLUSIONS: The present research provides a robust and efficient DL-based end-to-end PCG signal classification framework for designing a automated high-performance VHD diagnosis system.

Deep learning-accelerated computational framework based on Physics Informed Neural Network for the solution of linear elasticity.

The paper presents an efficient and robust data-driven deep learning (DL) computational framework developed for linear continuum elasticity problems. The methodology is based on the fundamentals of the Physics Informed Neural Networks (PINNs). For an accurate representation of the field variables, a multi-objective loss function is proposed. It consists of terms corresponding to the residual of the governing partial differential equations (PDE), constitutive relations derived from the governing physics, various boundary conditions, and data-driven physical knowledge fitting terms across randomly selected collocation points in the problem domain. To this end, multiple densely connected independent artificial neural networks (ANNs), each approximating a field variable, are trained to obtain accurate solutions. Several benchmark problems including the Airy solution to elasticity and the Kirchhoff-Love plate problem are solved. Performance in terms of accuracy and robustness illustrates the superiority of the current framework showing excellent agreement with analytical solutions. The present work combines the benefits of the classical methods depending on the physical information available in analytical relations with the superior capabilities of the DL techniques in the data-driven construction of lightweight, yet accurate and robust neural networks. The models developed herein can significantly boost computational speed using minimal network parameters with easy adaptability in different computational platforms.

Automated detection of lung nodules in CT scans: false-positive reduction with the radial-gradient index.

We present a number of approaches based on the radial gradient index (RGI) to achieve false-positive reduction in automated CT lung nodule detection. A database of 38 cases was used that contained a total of 82 lung nodules. For each CT section, a complementary image known as an "RGI map" was constructed to enhance regions of high circularity and thus improve the contrast between nodules and normal anatomy. Thresholds on three RGI parameters were varied to construct RGI filters that sensitively eliminated false-positive structures. In a consistency approach, RGI filtering eliminated 36% of the false-positive structures detected by the automated method without the loss of any true positives. Use of an RGI filter prior to a linear discriminant classifier yielded notable improvements in performance, with the false-positive rate at a sensitivity of 70% being reduced from 0.5 to 0.28 per section. Finally, the performance of the linear discriminant classifier was evaluated with RGI-based features. RGI-based features achieved a substantial improvement in overall performance, with a 94.8% reduction in the false-positive rate at a fixed sensitivity of 70%. These results demonstrate the potential role of RGI analysis in an automated lung nodule detection method.

Evaluation of automated lung nodule detection on low-dose computed tomography scans from a lung cancer screening program(1).

RATIONALE AND OBJECTIVES: The purpose of this study was to evaluate the performance of a fully automated lung nodule detection method in a large database of low-dose computed tomography (CT) scans from a lung cancer screening program. Because nodules demonstrate a spectrum of radiologic appearances, the performance of the automated method was evaluated on the basis of nodule malignancy status, size, subtlety, and radiographic opacity. MATERIALS AND METHODS: A database of 393 thick-section (10 mm) low-dose CT scans was collected. Automated lung nodule detection proceeds in two phases: gray-level thresholding for the initial identification of nodule candidates, followed by the application of a rule-based classifier and linear discriminant analysis to distinguish between candidates that correspond to actual lung nodules and candidates that correspond to non-nodules. Free-response receiver operating characteristic analysis was used to evaluate the performance of the method based on a jackknife training/testing approach. RESULTS: An overall nodule detection sensitivity of 70% (330 of 470) was attained with an average of 1.6 false-positive detections per section. At the same false-positive rate, 83% (57 of 69) of the malignant lung nodules in the database were detected. When the method was trained specifically for malignant nodules, a sensitivity of 80% (55 of 69) was attained with 0.85 false-positives per section. CONCLUSION: We have evaluated an automated lung nodule detection method with a large number of low-dose CT scans from a lung cancer screening program. An overall sensitivity of 80% for malignant nodules was achieved with 0.85 false-positive detections per section. Such a computerized lung nodule detection method is expected to become an important part of CT-based lung cancer screening programs.

A method for accurate spatial registration of PET images and histopathology slices.

BACKGROUND: Accurate alignment between histopathology slices and positron emission tomography (PET) images is important for radiopharmaceutical validation studies. Limited data is available on the registration accuracy that can be achieved between PET and histopathology slices acquired under routine pathology conditions where slices may be non-parallel, non-contiguously cut and of standard block size. The purpose of this study was to demonstrate a method for aligning PET images and histopathology slices acquired from patients with laryngeal cancer and to assess the registration accuracy obtained under these conditions. METHODS: Six subjects with laryngeal cancer underwent a (64)Cu-copper-II-diacetyl-bis(N4-methylthiosemicarbazone) ((64)Cu-ATSM) PET computed tomography (CT) scan prior to total laryngectomy. Sea urchin spines were inserted into the pathology specimen to act as fiducial markers. The specimen was fixed in formalin, as per standard histopathology operating procedures, and was then CT scanned and cut into millimetre-thick tissue slices. A subset of the tissue slices that included both tumour and fiducial markers was taken and embedded in paraffin blocks. Subsequently, microtome sectioning and haematoxylin and eosin staining were performed to produce 5-µm-thick tissue sections for microscopic digitisation. A series of rigid registration procedures was performed between the different imaging modalities (PET; in vivo CT-i.e. the CT component of the PET-CT; ex vivo CT; histology slices) with the ex vivo CT serving as the reference image. In vivo and ex vivo CTs were registered using landmark-based registration. Histopathology and ex vivo CT images were aligned using the sea urchin spines with additional anatomical landmarks where available. Registration errors were estimated using a leave-one-out strategy for in vivo to ex vivo CT and were estimated from the RMS landmark accuracy for histopathology to ex vivo CT. RESULTS: The mean ± SD accuracy for registration of the in vivo to ex vivo CT images was 2.66 ± 0.66 mm, and the accuracy for registration of histopathology to ex vivo CT was 0.86 ± 0.41 mm. Estimating the PET to in vivo CT registration accuracy to equal the PET-CT alignment accuracy of 1 mm resulted in an overall average registration error between PET and histopathology slices of 3.0 ± 0.7 mm. CONCLUSIONS: We have developed a registration method to align PET images and histopathology slices with an accuracy comparable to the spatial resolution of the PET images.

Automated lung nodule classification following automated nodule detection on CT: a serial approach.

We have evaluated the performance of an automated classifier applied to the task of differentiating malignant and benign lung nodules in low-dose helical computed tomography (CT) scans acquired as part of a lung cancer screening program. The nodules classified in this manner were initially identified by our automated lung nodule detection method, so that the output of automated lung nodule detection was used as input to automated lung nodule classification. This study begins to narrow the distinction between the "detection task" and the "classification task." Automated lung nodule detection is based on two- and three-dimensional analyses of the CT image data. Gray-level-thresholding techniques are used to identify initial lung nodule candidates, for which morphological and gray-level features are computed. A rule-based approach is applied to reduce the number of nodule candidates that correspond to non-nodules, and the features of remaining candidates are merged through linear discriminant analysis to obtain final detection results. Automated lung nodule classification merges the features of the lung nodule candidates identified by the detection algorithm that correspond to actual nodules through another linear discriminant classifier to distinguish between malignant and benign nodules. The automated classification method was applied to the computerized detection results obtained from a database of 393 low-dose thoracic CT scans containing 470 confirmed lung nodules (69 malignant and 401 benign nodules). Receiver operating characteristic (ROC) analysis was used to evaluate the ability of the classifier to differentiate between nodule candidates that correspond to malignant nodules and nodule candidates that correspond to benign lesions. The area under the ROC curve for this classification task attained a value of 0.79 during a leave-one-out evaluation.

Are pretreatment 18F-FDG PET tumor textural features in non-small cell lung cancer associated with response and survival after chemoradiotherapy?

UNLABELLED: There is evidence in some solid tumors that textural features of tumoral uptake in (18)F-FDG PET images are associated with response to chemoradiotherapy and survival. We have investigated whether a similar relationship exists in non-small cell lung cancer (NSCLC). METHODS: Fifty-three patients (mean age, 65.8 y; 31 men, 22 women) with NSCLC treated with chemoradiotherapy underwent pretreatment (18)F-FDG PET/CT scans. Response was assessed by CT Response Evaluation Criteria in Solid Tumors (RECIST) at 12 wk. Overall survival (OS), progression-free survival (PFS), and local PFS (LPFS) were recorded. Primary tumor texture was measured by the parameters coarseness, contrast, busyness, and complexity. The following parameters were also derived from the PET data: primary tumor standardized uptake values (SUVs) (mean SUV, maximum SUV, and peak SUV), metabolic tumor volume, and total lesion glycolysis. RESULTS: Compared with nonresponders, RECIST responders showed lower coarseness (mean, 0.012 vs. 0.027; P = 0.004) and higher contrast (mean, 0.11 vs. 0.044; P = 0.002) and busyness (mean, 0.76 vs. 0.37; P = 0.027). Neither complexity nor any of the SUV parameters predicted RECIST response. By Kaplan-Meier analysis, OS, PFS, and LPFS were lower in patients with high primary tumor coarseness (median, 21.1 mo vs. not reached, P = 0.003; 12.6 vs. 25.8 mo, P = 0.002; and 12.9 vs. 20.5 mo, P = 0.016, respectively). Tumor coarseness was an independent predictor of OS on multivariable analysis. Contrast and busyness did not show significant associations with OS (P = 0.075 and 0.059, respectively), but PFS and LPFS were longer in patients with high levels of each (for contrast: median of 20.5 vs. 12.6 mo, P = 0.015, and median not reached vs. 24 mo, P = 0.02; and for busyness: median of 20.5 vs. 12.6 mo, P = 0.01, and median not reached vs. 24 mo, P = 0.006). Neither complexity nor any of the SUV parameters showed significant associations with the survival parameters. CONCLUSION: In NSCLC, baseline (18)F-FDG PET scan uptake showing abnormal texture as measured by coarseness, contrast, and busyness is associated with nonresponse to chemoradiotherapy by RECIST and with poorer prognosis. Measurement of tumor metabolic heterogeneity with these parameters may provide indices that can be used to stratify patients in clinical trials for lung cancer chemoradiotherapy.

Deformable density matching for 3D non-rigid registration of shapes.

There exists a large body of literature on shape matching and registration in medical image analysis. However, most of the previous work is focused on matching particular sets of features--point-sets, lines, curves and surfaces. In this work, we forsake specific geometric shape representations and instead seek probabilistic representations--specifically Gaussian mixture models--of shapes. We evaluate a closed-form distance between two probabilistic shape representations for the general case where the mixture models differ in variance and the number of components. We then cast non-rigid registration as a deformable density matching problem. In our approach, we take one mixture density onto another by deforming the component centroids via a thin-plate spline (TPS) and also minimizing the distance with respect to the variance parameters. We validate our approach on synthetic and 3D arterial tree data and evaluate it on 3D hippocampal shapes.