Convolutional Neural Networks to Classify Alzheimer’s Disease Severity Based on SPECT Images: A Comparative Study

Automatic prediction of citrus fruit diseases requires the selection of a suitable classification model among various Convolutional Neural Network (CNN) models. The CNN models are the most effective for image classification. However, more than one CNN model occasionally performs closer to each other. Choosing the best classification model requires considerable effort in this situation. In this study, 17 CNN models were used for citrus fruit disease identification. Citrus fruit diseases include anthracnose, blackspot, canker, scab, and healthy. In this research, the performance of CNN models was analysed, and a two-phase statistical analysis was carried out to determine which CNN model provided the most accurate predictions of citrus crop disease. The 17 CNN models used were AlexNet, Darknet, Densenet, efficientnet, nasnetlarge, googlenet, inceptionresnetv2, inceptionv3, mobilenetv2, nasnetmobile, Xception, squeezenet, resnet18, resnet50, resnet101, vgg16, and vgg19. The two-phase statistical analysis was Duncan's multiple range test, followed by the Wilcoxon signed-rank test is the second one. Again, in order to do a statistical analysis, nine different indicators, including “accuracy, sensitivity, specificity, precision, FPR, F1 Score, MCC, Kappa, and computing time”, were taken into consideration. After the execution of 17 CNN models and twophase statistical analysis, it was revealed that resnet101 is superior among CNN models for citrus fruit disease prediction.

PurposeSemantic segmentation is a fundamental part of the surgical application of deep learning. Traditionally, segmentation in vision tasks has been performed using convolutional neural networks (CNNs), but the transformer architecture has recently been introduced and widely investigated. We aimed to investigate the performance of deep learning models in segmentation in robot-assisted radical prostatectomy (RARP) and identify which of the architectures is superior for segmentation in robotic surgery.Materials and MethodsIntraoperative images during RARP were obtained. The dataset was randomly split into training and validation data. Segmentation of the surgical instruments, bladder, prostate, vas and seminal vesicle was performed using three CNN models (DeepLabv3, MANet, and U-Net++) and three transformers (SegFormer, BEiT, and DPT), and their performances were analyzed.ResultsThe overall segmentation performance during RARP varied across different model architectures. For the CNN models, DeepLabV3 achieved a mean Dice score of 0.938, MANet scored 0.944, and U-Net++ reached 0.930. For the transformer architectures, SegFormer attained a mean Dice score of 0.919, BEiT scored 0.916, and DPT achieved 0.940. The performance of CNN models was superior to that of transformer models in segmenting the prostate, vas, and seminal vesicle.ConclusionsDeep learning models provided accurate segmentation of the surgical instruments and anatomical structures observed during RARP. Both CNN and transformer models showed reliable predictions in the segmentation task; however, CNN models may be more suitable than transformer models for organ segmentation and may be more applicable in unusual cases. Further research with large datasets is needed.

The worldwide population and the healthcare industry have been put at risk by the coronavirus pandemic .Distinguishing COVID-19 affected individuals from healthy patient groups is an important job.Rapid and efficient screening methods are desperately needed in the battle against this unique illness in order to detect patients at an early stage and classify findings objectively.In order to identify COVID-19 from Chest X-ray (CXR) pictures, this research improved a robust Convolutional Neural Network (CNN) model that was constructed, called DeepCovNet, using the Synthetic Minority Over-sampling Technique (SMOTE).The principal issue tackled was the inherent class disparity in medical imaging datasets, wherein patients positive for COVID-19 are frequently underestimated in comparison to normal cases.The CNN model's performance was assessed after it was trained on a balanced dataset created via SMOTE.The improved method performs better than the current one with accuracy 96.0%, classifying COVID-19 patient cases using CXR images.The result of this study indicates a significant development in the healthcare industry and society at large in the battle against the pandemic employing cutting-edge deep learning techniques.This notable enhancement is related to the integration of (SMOTE).This study suggests that in order to improve model performance in scenarios with more severe class imbalance cases, additional research should investigate the combination of SMOTE with other data augmentation techniques, and this model should be extended to other medical imaging tasks, such as the detection of different lung diseases or pneumonias.

Convolutional Neural Networks (CNNs) have demonstrated impressive performance in complex machine learning tasks such as classification and regression problems. A reliable neural network structure plays a decisive role in CNN studies. Through comparing and analyzing the structure of neural networks, a model structure for better visualization performance has been discovered, and such a method supports the development of deep learning research. These studies are of particular importance in end-to-end systems for autonomous driving to imitate human driving, where the interpretability of the system is limited. Because of the uncertainty of the ground truth, for the determination of human steering in an image, it is difficult to accurately compare the visualization performance of different CNN models or different visualization methods. For practical applications, however, an objective and quantitative measure for assessing visualization performance is necessary. Therefore, a method to evaluate the visualization performance of CNN models using a driver model instead of human drivers is proposed, to generate a data set which can be used to determine the decisional point (ground truth) in the input image. Then, an exclusive method is also put forth, to quantitatively calculate the relationship between the decisional point (ground truth) and the visualization results produced by CNN models. In this paper, five CNN models as an autonomous steering controller are designed based on PilotNet, and the visualization abilities of each CNN models is compared by three evaluation indicators. By comparing the visualization performance of five different CNN models, it is shown that the proposed method can successfully assess the visualization level of the CNN model.

ObjectivesTo analyse the differences in the clinical features and characteristics of 123I-labelled 2β-carbomethoxy-3β-(4-iodophenyl)-N-(3-fluoropropyl) nortropane (123I-FP-CIT) single photon emission CT (SPECT) imaging among patients with vascular parkinsonism (VP) and Parkinson's disease...

Brain perfusion single photon emission computed tomography (SPECT) scans were initially developed in 1970's. A key radiopharmaceutical, hexamethylpropyleneamine oxime (HMPAO), was originally approved in 1988, but was unstable. As a result, the quality of SPECT images varied greatly based on technique until 1993, when a method of stabilizing HMPAO was developed. In addition, most SPECT perfusion studies pre-1996 were performed on single-head gamma cameras. In 1996, the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology (TTASAAN) issued a report regarding the use of SPECT in the evaluation of neurological disorders. Although the TTASAAN report was published in January 1996, it was approved for publication in October 1994. Consequently, the reported brain SPECT studies relied upon to derive the conclusions of the TTASAAN report largely pre-date the introduction of stabilized HMPAO. While only 12% of the studies on traumatic brain injury (TBI) in the TTASAAN report utilized stable tracers and multi-head cameras, 69 subsequent studies with more than 23,000 subjects describe the utility of perfusion SPECT scans in the evaluation of TBI. Similarly, dementia SPECT imaging has improved. Modern SPECT utilizing multi-headed gamma cameras and quantitative analysis has a sensitivity of 86% and a specificity of 89% for the diagnosis of mild to moderate Alzheimer's disease—comparable to fluorodeoxyglucose positron emission tomography. Advances also have occurred in seizure neuroimaging. Lastly, developments in SPECT imaging of neurotoxicity and neuropsychiatric disorders have been striking. At the 25-year anniversary of the publication of the TTASAAN report, it is time to re-examine the utility of perfusion SPECT brain imaging. Herein, we review studies cited by the TTASAAN report vs. current brain SPECT imaging research literature for the major indications addressed in the report, as well as for emerging indications. In Part II, we elaborate technical aspects of SPECT neuroimaging and discuss scan interpretation for the clinician.

Understanding the learning mechanism of convolutional neural networks in spectral analysis

We aimed to evaluate the image quality and diagnostic performance of chronic lung allograft dysfunction (CLAD) with lung ventilation single-photon emission computed tomography (SPECT) images acquired briefly using a convolutional neural network (CNN) in patients after lung transplantation and to explore the feasibility of short acquisition times. We retrospectively identified 93 consecutive lung-transplant recipients who underwent ventilation SPECT/computed tomography (CT). We employed a CNN to distinguish the images acquired in full time from those acquired in a short time. The image quality was evaluated using the structural similarity index (SSIM) loss and normalized mean square error (NMSE). The correlation between functional volume/morphological volume (F/M) ratios of full-time SPECT images and predicted SPECT images was evaluated. Differences in the F/M ratio were evaluated using Bland-Altman plots, and the diagnostic performance was compared using the area under the curve (AUC). The learning curve, obtained using MSE, converged within 100 epochs. The NMSE was significantly lower (P < 0.001) and the SSIM was significantly higher (P < 0.001) for the CNN-predicted SPECT images compared to the short-time SPECT images. The F/M ratio of full-time SPECT images and predicted SPECT images showed a significant correlation (r = 0.955, P < 0.0001). The Bland-Altman plot revealed a bias of -7.90% in the F/M ratio. The AUC values were 0.942 for full-time SPECT images, 0.934 for predicted SPECT images and 0.872 for short-time SPECT images. Our findings suggest that a deep-learning-based approach can significantly curtail the acquisition time of ventilation SPECT, while preserving the image quality and diagnostic accuracy for CLAD.

Convolutional neural network (CNN) is a promising image recognition technique for winter road surface condition (RSC), a measure that is crucial for winter maintenance operations. In the past, researchers have designed RSC CNN models that displayed acceptable results but did so focusing solely on obtaining high classification accuracy without any consideration for efficiency. Furthermore, when it comes to model development itself, architecture design requires expertise in CNN as well as rich knowledge in the investigated problem itself. To rectify these issues, this paper proposes an innovative approach to automatically design RSC CNN architecture without compromising classification accuracy. The proposed approach uses a weighted sum method, which provides the freedom of choosing relative importance level between accuracy and efficiency. Once the relative importance has been set, one of the most successful and widely adopted heuristics, namely, simulated annealing (SA), is employed to generate (sub)optimal solutions. Results show that both accuracy and efficiency of the automatically generated CNNs are better or at least comparable to the two selected state-of-the-art CNN models, ResNet50 and MobileNet, achieving as high as 93.44% classification accuracy. Ultimately, the outcome of this study fills the gap in existing CNN design methods that do not consider the tradeoff between accuracy and efficiency while providing insight into the effect varying architectures have on CNN model performance.

Low spatial resolution and the related partial volume effects limit the diagnostic potential of brain single photon emission computed tomography (SPECT) imaging. As a possible remedy for this problem we propose a technique for the fusion of SPECT and MR images, which requires for a given patient the SPECT data and the T1-weighted MR image. Basically, after the reconstruction and coregistration steps, the high-frequency part of the MR, which would be unrecoverable by the set SPECT acquisition system + reconstruction algorithm, is extracted and added to the SPECT image. The tuning of the weight of the MR on the resulting fused image can be performed very quickly, any iterative reconstruction algorithm can be used and, in the case that the SPECT projections are not available, the proposed technique can also be applied directly to the SPECT image, provided that the performance of the scanner is known. The procedure has the potential of increasing the diagnostic value of a SPECT image. Even in the locations of SPECT-MR mismatch it does not significantly affect quantitation over regions of interest (ROIs) whose dimensions are decidedly larger than the SPECT resolution distance. On the other hand, appreciable corrections for partial volume effects are expected in the locations where the contrast in the structural MR matches the corresponding contrast in functional activity.

This study leveraged Convolutional Neural Network (CNN) models to estimate canopy height in Southeast Australian forests before and after the 2019–2020 bushfire event, using inputs from Sentinel-1, Sentinel-2, spectral indices, and terrain features and GEDI canopy height (rh99) as target variables. Our research consisted of three primary objectives: (1) an evaluation of GEDI rh99 canopy height in relation to an nDSM derived from airborne LiDAR, (2) the development of two CNN models for pre- and post-fire scenarios, and (3) utilizing the trained CNN models to generate pre- and post-fire canopy height maps and canopy height change maps across the continuous landscape. GEDI rh99 accuracy analysis revealed R 2 = 0.85, RMSE = 7.25 m and nRMSE = 0.23 for pre-fire GEDI and R 2 = 0.78, RMSE = 10.1 m and nRMSE = 0.3 for post-fire GEDI. The pre-fire CNN model achieved R 2 = 0.75, RMSE = 7.29 m and nRMSE = 0.22, and post-fire CNN model achieved R 2 = 0.62, RMSE = 11.67 m, and nRMSE = 0.34 when validated against nDSM. Bias analysis was conducted, revealing that GEDI rh99 underestimated nDSM across all canopy height ranges, and slope has no significant impact on GEDI accuracy. Notably, post-fire GEDI rh99 exhibited significant underestimation in high to extreme fire severity. The CNN models’ prediction displayed overestimation in short forests < 20 m and increasing underestimation in taller forests > 20 m and underestimation in high and extreme fire severity. Subsequent investigation indicated that canopy layer thinning in high fire severity conditions resulted in weak GEDI waveform signals and consequent bias. Additionally, the GEDI rh99 bias was propagated to the CNN models, together with insensitivity of spectral band values to canopy height in dense and burnt forests potentially contributed to models’ bias. Finally, we demonstrated CNN models’ ability to monitor forest recovery by generating canopy height maps spanning East Gippsland from 2019 to 2023.

Plasma soluble neuregulin-1 as a diagnostic biomarker for Alzheimer's disease

to assess diagnostic capabilities of single-photon emission computed tomography (SPECT) in the detection of obstructive coronary artery (CA) lesions, depending on the meeting appropriate use criteria. We used in this retrospective analysis data from 107 patients with previously diagnosed ischemic heart disease (IHD) or in need to exclude it, who were hospitalized in inpatient departments of the Research Institute for Complex Issues of Cardiovascular Diseases in the period from 2012 to 2015. All patients underwent coronary angiography (CAG) and SPECT (the time interval between the studies did not exceed 3 months) for detection of hemodynamically significant CA stenoses. Patients were distributed into two groups according SPECT imaging appropriateness score: group 1-88 patients with score 7-9 (in whom SPECT imaging was appropriate), group 2-19 patients with score 1-6 (in whom SPECT imaging was uncertain, possibly appropriate, or inappropriate. Clinical signs and symptoms of angina pectoris were predominantly found in group 1 patients (p=0.499). Asymptomatic patients were more likely to be found in group 2 (p&lt;0.001). Group 1 patients commonly had high pretest probability (PTP) (over 90 %, p&lt;0.001), whereas group 2 patients commonly had low PTP (5-10 %, p&lt;0.001). Mean PTP was 77 and 58 % in groups 1 and 2, respectively (p=0.003). According to positive SPECT imaging, significant CA lesions were more often found in group 1 compared to group 2 (31.8 and 10.5 %, respectively, p=0.060). Two- and three-vessel disease prevailed in group 1 (25 % and 14.7 %) according the analysis of prevalence and location of hemodynamically significant CA lesions, although the data did not reach statistical significance (p=0.057 and p=0.073). Stenoses &gt;70 % were more commonly detected in group 1, compared to group 2: in anterior descending artery 52.3 vs. 5.3 % (p&lt;0.001), circumflex artery 35.2 vs. 10.5 %; (p=0.034), right coronary artery 34.1 vs. 10.5 % (p=0.041). The sensitivity in both groups was rather low (40 % vs. 25 %), whereas specificity was 83 % in group 1 and 93 % in group 2. According to clinical examination, patients with IHD and indications for SPECT imaging more often had obstructive CA lesions (63.6 %), than patients with questionable or no indications (21.1 %). However, rate of positive findings during stress tests with SPECT imaging was low in both groups and did not differ significantly (p=0.06). Despite high specificity of SPECT imaging, its sensitivity was low in both groups.

Multimodality nuclear medicine imaging in three-dimensional radiation treatment planning for lung cancer: challenges and prospects

To better understand and improve the prediction of the seasonal surface temperature (TS) across the United States, we employed convolutional neural network (CNN) models trained on the Community Earth System Model Version 2 Large Ensemble (CESM2 LENS). We used lagged sea surface temperatures (SST) over the tropical Pacific region, containing the information of the El Niño Southern Oscillation (ENSO), as input for the CNN models. ENSO is the principal driver of variability in seasonal US surface temperatures (TSUS) and employing CNN models allows for spatiotemporal aspects of ENSO to be analyzed to make seasonal TSUS predictions. For predicting TSUS, the CNN models exhibited significantly improved skill over standard statistical multilinear regression (MLR) models and dynamical forecasts across most regions in the US, for lead times ranging from 1 to 6 months. Furthermore, we employed the CNN models to predict seasonal TSUS during extreme ENSO events. For these events, the CNN models outperformed the MLR models in predicting the effects on seasonal TSUS, suggesting that the CNN models are able to capture the ENSO‐TSUS teleconnection more effectively. Results from a heatmap analysis demonstrate that the CNN models utilize spatial features of ENSO rather than solely the magnitude of the ENSO events, indicating that the improved skill of seasonal TSUS is due to analyzing spatial variation in ENSO events. The proposed CNN model demonstrates a promising improvement in prediction skill compared to existing methods, suggesting a potential path forward for enhancing TSUS forecast skill from subseasonal to seasonal timescales.