A multi-class brain tumor grading system based on histopathological images using a hybrid YOLO and RESNET networks

Gliomas are primary brain tumors caused by glial cells. These cancers' classification and grading are crucial for prognosis and treatment planning. Deep learning (DL) can potentially improve the digital pathology investigation of brain tumors. In this paper, we developed a technique for visualizing a predictive tumor grading model on histopathology pictures to help guide doctors by emphasizing characteristics and heterogeneity in forecasts. The proposed technique is a hybrid model based on YOLOv5 and ResNet50. The function of YOLOv5 is to localize and classify the tumor in large histopathological whole slide images (WSIs). The suggested technique incorporates ResNet into the feature extraction of the YOLOv5 framework, and the detection results show that our hybrid network is effective for identifying brain tumors from histopathological images. Next, we estimate the glioma grades using the extreme gradient boosting classifier. The high-dimensional characteristics and nonlinear interactions present in histopathology images are well-handled by this classifier. DL techniques have been used in previous computer-aided diagnosis systems for brain tumor diagnosis. However, by combining the YOLOv5 and ResNet50 architectures into a hybrid model specifically designed for accurate tumor localization and predictive grading within histopathological WSIs, our study presents a new approach that advances the field. By utilizing the advantages of both models, this creative integration goes beyond traditional techniques to produce improved tumor localization accuracy and thorough feature extraction. Additionally, our method ensures stable training dynamics and strong model performance by integrating ResNet50 into the YOLOv5 framework, addressing concerns about gradient explosion. The proposed technique is tested using the cancer genome atlas dataset. During the experiments, our model outperforms the other standard ways on the same dataset. Our results indicate that the proposed hybrid model substantially impacts tumor subtype discrimination between low-grade glioma (LGG) II and LGG III. With 97.2% of accuracy, 97.8% of precision, 98.6% of sensitivity, and the Dice similarity coefficient of 97%, the proposed model performs well in classifying four grades. These results outperform current approaches for identifying LGG from high-grade glioma and provide competitive performance in classifying four categories of glioma in the literature.

A multi-resolution TOF-DOI detector for human brain dedicated PET scanner

Objective. We propose a single-ended readout, multi-resolution detector design that can achieve high spatial, depth-of-interaction (DOI), and time-of-flight (TOF) resolutions, as well as high sensitivity for human brain-dedicated positron emission tomography (PET) scanners.Approach. The detector comprised two layers of LYSO crystal arrays and a lightguide in between. The top (gamma ray entrance) layer consisted of a 16 × 16 array of 1.53 × 1.53 × 6 mm3LYSO crystals for providing high spatial resolution. The bottom layer consisted of an 8 × 8 array of 3.0 × 3.0 × 15 mm3LYSO crystals that were one-to-one coupled to an 8 × 8 multipixel photon counter (MPPC) array for providing high TOF resolution. The 2 mm thick lightguide introduces inter-crystal light sharing that causes variations of the light distribution patterns for high DOI resolution. The detector was read out by a PETsys TOFPET2 application-specific integrated circuit.Main result. The top and bottom layers were distinguished by a convolutional neural network with 97% accuracy. All crystals in the top and bottom layers were resolved. The inter-crystal scatter (ICS) events in the bottom layer were identified, and the measured average DOI resolution of the bottom layer was 4.1 mm. The coincidence time resolution (CTR) for the top-top, top-bottom, and bottom-bottom coincidences was 476 ps, 405 ps, and 298 ps, respectively. When ICS events were excluded from the bottom layer, the CTR of the bottom-bottom coincidence was 277 ps.Significance. The top layer of the proposed two-layer detector achieved a high spatial resolution and the bottom layer achieved a high TOF resolution. Together with its high DOI resolution and detection efficiency, the proposed detector is well suited for next-generation high-performance brain-dedicated PET scanners.

Development and evaluation of a new high-TOF-resolution all-digital brain PET system

Objective.Time-of-flight (TOF) capability and high sensitivity are essential for brain-dedicated positron emission tomography (PET) imaging, as they improve the contrast and the signal-to-noise ratio (SNR) enabling a precise localization of functional mechanisms in the different brain regions.Approach.We present a new brain PET system with transverse and axial field-of-view (FOV) of 320 mm and 255 mm, respectively. The system head is an array of 6 × 6 detection elements, each consisting of a 3.9 × 3.9 × 20 mm3lutetium-yttrium oxyorthosilicate crystal coupled with a 3.93 × 3.93 mm2SiPM. The SiPMs analog signals are individually digitized using the multi-voltage threshold (MVT) technology, employing a 1:1:1 coupling configuration.Main results.The brain PET system exhibits a TOF resolution of 249 ps at 5.3 kBq ml-1, an average sensitivity of 22.1 cps kBq-1, and a noise equivalent count rate (NECR) peak of 150.9 kcps at 8.36 kBq ml-1. Furthermore, the mini-Derenzo phantom study demonstrated the system's ability to distinguish rods with a diameter of 2.0 mm. Moreover, incorporating the TOF reconstruction algorithm in an image quality phantom study optimizes the background variability, resulting in reductions ranging from 44% (37 mm) to 75% (10 mm) with comparable contrast. In the human brain imaging study, the SNR improved by a factor of 1.7 with the inclusion of TOF, increasing from 27.07 to 46.05. Time-dynamic human brain imaging was performed, showing the distinctive traits of cortex and thalamus uptake, as well as of the arterial and venous flow with 2 s per time frame.Significance.The system exhibited a good TOF capability, which is coupled with the high sensitivity and count rate performance based on the MVT digital sampling technique. The developed TOF-enabled brain PET system opens the possibility of precise kinetic brain PET imaging, towards new quantitative predictive brain diagnostics.

Perspectives of brain imaging with PET systems

In this partial review and partial attempt at vision of what may be the future of dedicated brain PET scanners, the key implementations of the PET technique, we postulate that we are still on a development path and there is still a lot to be done in order to develop optimal brain imagers. Optimized for particular imaging tasks and protocols, and also mobile, that can be used outside the PET center, in addition to the expected improvements in sensitivity and resolution. For this multi-application concept to be more practical, flexible, adaptable designs are preferred. This task is greatly facilitated by the improved TOF performance that allows for more open, adjustable, limited angular coverage geometries without creating image artifacts. As achieving uniform very high resolution in the whole body is not practical due to technological limits and high costs, hybrid systems using a moderate-resolution total body scanner (such as J-PET) combined with a very high performing brain imager could be a very attractive approach. As well, as using magnification inserts in the total body or long-axial length imagers to visualize selected targets with higher resolution. In addition, multigamma imagers combining PET with Compton imaging should be developed to enable multitracer imaging.

Quantitative PET in the 2020s: a roadmap

Positron emission tomography (PET) plays an increasingly important role in research and clinical applications, catalysed by remarkable technical advances and a growing appreciation of the need for reliable, sensitive biomarkers of human function in health and disease. Over the last 30 years, a large amount of the physics and engineering effort in PET has been motivated by the dominant clinical application during that period, oncology. This has led to important developments such as PET/CT, whole-body PET, 3D PET, accelerated statistical image reconstruction, and time-of-flight PET. Despite impressive improvements in image quality as a result of these advances, the emphasis on static, semi-quantitative 'hot spot' imaging for oncologic applications has meant that the capability of PET to quantify biologically relevant parameters based on tracer kinetics has not been fully exploited. More recent advances, such as PET/MR and total-body PET, have opened up the ability to address a vast range of new research questions, from which a future expansion of applications and radiotracers appears highly likely. Many of these new applications and tracers will, at least initially, require quantitative analyses that more fully exploit the exquisite sensitivity of PET and the tracer principle on which it is based. It is also expected that they will require more sophisticated quantitative analysis methods than those that are currently available. At the same time, artificial intelligence is revolutionizing data analysis and impacting the relationship between the statistical quality of the acquired data and the information we can extract from the data. In this roadmap, leaders of the key sub-disciplines of the field identify the challenges and opportunities to be addressed over the next ten years that will enable PET to realise its full quantitative potential, initially in research laboratories and, ultimately, in clinical practice.

Deep Learning for Multigrade Brain Tumor Classification in Smart Healthcare Systems: A Prospective Survey

Brain tumor is one of the most dangerous cancers in people of all ages, and its grade recognition is a challenging problem for radiologists in health monitoring and automated diagnosis. Recently, numerous methods based on deep learning have been presented in the literature for brain tumor classification (BTC) in order to assist radiologists for a better diagnostic analysis. In this overview, we present an in-depth review of the surveys published so far and recent deep learning-based methods for BTC. Our survey covers the main steps of deep learning-based BTC methods, including preprocessing, features extraction, and classification, along with their achievements and limitations. We also investigate the state-of-the-art convolutional neural network models for BTC by performing extensive experiments using transfer learning with and without data augmentation. Furthermore, this overview describes available benchmark data sets used for the evaluation of BTC. Finally, this survey does not only look into the past literature on the topic but also steps on it to delve into the future of this area and enumerates some research directions that should be followed in the future, especially for personalized and smart healthcare.

Roadmap toward the 10 ps time-of-flight PET challenge

Since the seventies, positron emission tomography (PET) has become an invaluable medical molecular imaging modality with an unprecedented sensitivity at the picomolar level, especially for cancer diagnosis and the monitoring of its response to therapy. More recently, its combination with x-ray computed tomography (CT) or magnetic resonance (MR) has added high precision anatomic information in fused PET/CT and PET/MR images, thus compensating for the modest intrinsic spatial resolution of PET. Nevertheless, a number of medical challenges call for further improvements in PET sensitivity. These concern in particular new treatment opportunities in the context personalized (also called precision) medicine, such as the need to dynamically track a small number of cells in cancer immunotherapy or stem cells for tissue repair procedures. A better signal-to-noise ratio (SNR) in the image would allow detecting smaller size tumours together with a better staging of the patients, thus increasing the chances of putting cancer in complete remission. Moreover, there is an increasing demand for reducing the radioactive doses injected to the patients without impairing image quality. There are three ways to improve PET scanner sensitivity: improving detector efficiency, increasing geometrical acceptance of the imaging device and pushing the timing performance of the detectors. Currently, some pre-localization of the electron-positron annihilation along a line-of-response (LOR) given by the detection of a pair of annihilation photons is provided by the detection of the time difference between the two photons, also known as the time-of-flight (TOF) difference of the photons, whose accuracy is given by the coincidence time resolution (CTR). A CTR of about 10 picoseconds FWHM will ultimately allow to obtain a direct 3D volume representation of the activity distribution of a positron emitting radiopharmaceutical, at the millimetre level, thus introducing a quantum leap in PET imaging and quantification and fostering more frequent use of 11C radiopharmaceuticals. The present roadmap article toward the advent of 10 ps TOF-PET addresses the status and current/future challenges along the development of TOF-PET with the objective to reach this mythic 10 ps frontier that will open the door to real-time volume imaging virtually without tomographic inversion. The medical impact and prospects to achieve this technological revolution from the detection and image reconstruction point-of-views, together with a few perspectives beyond the TOF-PET application are discussed.

Quantifying bias and precision of kinetic parameter estimation on the PennPET Explorer, a long axial field-of-view scanner

Long axial field-of-view (AFOV) PET scanners allow for full-body dynamic imaging in a single bed-position at very high sensitivity. However, the benefits for kinetic parameter estimation have yet to be studied. This work uses (1) a dynamic GATE simulation of [¹⁸F]-fluorothymidine (FLT) in a modified NEMA IQ phantom and (2) a lesion embedding study of spheres in a dynamic [¹⁸F]-fluorodeoxyglucose (FDG) human subject imaged on the PennPET Explorer. Both studies were designed using published kinetic data of lung and liver cancers and modeled using two tissue compartments. Data were reconstructed at various emulated doses. Sphere time-activity curves (TACs) were measured on resulting dynamic images, and TACs were fit using a two-tissue-compartment model (k₄ ≠ 0) for the FLT study and both a two-tissue-compartment model (k₄ = 0) and Patlak graphical analysis for the FDG study to estimate flux (K_i) and delivery (K₁) parameters. Quantification of flux and K1 shows lower bias and better precision for both radiotracers on the long AFOV scanner, especially at low doses. Dynamic imaging on a long AFOV system can be achieved for a greater range of injected doses, as low as 0.5-2 mCi depending on the sphere size and flux, compared to a standard AFOV scanner, while maintaining good kinetic parameter estimation.

Design study of a dedicated head and neck cancer PET system

The tumor-involved regions of head and neck cancer (HNC) have complex anatomical structures and vital physiological roles. As a consequence, there is a need for high sensitivity and high spatial resolution dedicated HNC PET scanner. The purpose of this study is to evaluate and optimize system design that includes detecting materials and geometries. For the detecting material, two scanners with the same two-panel geometry based on CZT and LYSO were evaluated. For the system geometry, four CZT scanners with two-panel, lengthened two-panel, four-panel, and full-ring geometries were evaluated. A cylinder phantom with sphere lesions and an XCAT phantom in the head and neck region were simulated. The results showed that the sensitivity of the 40-mm thickness CZT system and the 20-mm thickness LYSO system were comparable. However, the multiple interaction photon events recovery accuracy of the CZT system was about 20% higher. The in-panel and orthogonal-panel spatial resolutions of CZT are 0.58 and 0.74 mm, while those of LYSO are 0.70 and 1.40 mm. For system geometry, the four-panel and full-ring scanners have a higher contrast recovery coefficient (CRC) and contrast-to-noise ratio (CNR) than the two-panel and lengthened two-panel scanners. However, a 5-mm lesion in the XCAT phantom was visualized within 6 min in the two-panel system.