Outcomes of Tolerability, Acute Toxicity and Quality of Life from MR-Guided Radiation Therapy (1.5T MR-Linac) for Liver Metastases in the MOMENTUM Study

PURPOSE/OBJECTIVE(S)

Stereotactic body radiation therapy (SBRT) offers an important treatment option for metastatic liver tumors. The introduction of magnetic resonance (MR) guided SBRT has paved the way for optimal tumor visualization and daily plan adaptation. The purpose of this study is to review tolerability of MR-guided SBRT of liver metastases and to present early toxicity and quality of life outcomes.

MATERIALS/METHODS

All patients with liver metastases who were treated on a 1.5T MR-Linac and enrolled in the MOMENTUM study (NTC04075305) were included. Patients were treated between April 2019 and December 2022 in 5 different institutes across 3 countries. Descriptive statistics were used to present the tolerability of treatment, toxicity (CTCAE v5.0) and quality of life outcomes (QLQ C30 and EQ 5D-5L) at baseline and 3 months after treatment.

RESULTS

A total of 127 patients with liver metastases were included in the analysis. There were 64 females and 63 men, with a median age of 66 years (range 31 to 93). The median ECOG-score was 0 (range 0-2). The most common primary origin was colorectal cancer (66%), followed by bronchus and lung cancer (12%), with ocular melanoma, pancreatic and breast cancer being joint third (all 6-7%). Fractionation schedules ranged from 12 - 67.5 Gy in 2 - 12 fractions. The most commonly prescribed fractionation dose was 60 Gy in 3-5 fractions (53% and 13% respectively) and 50 Gy in 5 fractions (11%). Completion data was available for 116 patients. 112 patients (97%) received all fractions. 4 Patients (3%) did not complete treatment due to technical issues and 2 of the 4 receiving no treatment on the MR-Linac. Physician reported toxicity at 3 months was recorded for 82 patients (66%). No grade 4 or 5 toxicities were reported. There were 12 grade 3 toxicities reported in 6 (7%) patients with 5 deemed radiation therapy related (Table 1) and 34 grade 2 toxicities in 21 (26%) patients.

CONCLUSION

We have presented the largest cohort (to our knowledge) of 127 patients treated using 1.5 Tesla MR Guidance for metastatic liver tumors. 97% of treatments were completed successfully with all treatments being well tolerated. Acute grade 3 toxicity was reported in 7% of patients with no grade 4 or 5 toxicities present. These outcomes suggest radiotherapy on the MR-Linac is a safe and promising treatment for patients with liver metastases. Additional prospective follow up is ongoing for late toxicity events and long-term control data. Table 1: Grade ≥3 toxicity at 3 months related to radiation therapy (total No. of patients was 82) There was QLQ-C30 data on 89 patients at baseline and on 62 patients at 3 months. At 3 months the median score was worse for physical functioning, VAS score and pain.

Differential Response in Quantitative MRI Parameters Detected in Head and Neck Cancer Patients Treated with Concurrent Immunotherapy during Hypo-Fractionated MR-gRT

PURPOSE/OBJECTIVE(S)

Recently, multiple clinical trials have demonstrated success of PD-1/PD-L1-targeted immune checkpoint inhibition (ICI) in recurrent and metastatic head and neck cancer (HNC). However, three large clinical trials combining PD-1/PD-L1-targeted ICI with RT or chemo-RT (CRT) in the definitive management of HNC have shown no benefit of combination therapy with ICI over RT or CRT alone. Our overarching hypotheses are: i) hypo-fractionation may ultimately better synergize with ICI compared to conventional RT regimens, and ii) immunologic changes in the tumor microenvironment may be detectable using quantitative MRI (qMRI) parameters collected during RT.

MATERIALS/METHODS

Seven patients treated with hypo-fractionated MR-guided RT of 50 Gy in 15 fractions (DEHART, NCT04477759) were included in the study. Four patients (Group 1) were treated with concurrent atezolizumab (a monoclonal antibody) and three patients (Group 2) were treated with RT alone. Daily DWI, T1 mapping, and T2 mapping sequences were acquired on a 1.5T MR-Linac in the idle time during adaptive plan generation. Median ADC, T1, T2, and Dslow (derived from b-values 150 and 550 s/mm2) values were extracted from physician-defined GTV and manually constructed posterior paraspinal muscle contours, the latter serving as a control. Wilcoxon signed rank tests were conducted using pre/post treatment data for each qMRI parameter.

RESULTS

GTV ADC, Dslow, T2, and T1 increased for both patient groups over the course of treatment with significant differences in ADC, Dslow, and T2 detected between fractions 1 and 15 for all patients studied (p = 0.0156, p = 0.0156, and p = 0.0469, respectively). No significant differences were detected in control qMRI parameters pre/post treatment. No significant differences in ADC, Dslow, and T2 were detected between groups' fractions 1 and 15 in these small cohorts. However, interestingly, we observed a differential change in the increase of median GTV T2 and Dslow values during fractions 10-12 in Group 1 compared to Group 2, suggesting this time interval may prime the anti-tumor immune response.

CONCLUSION

Combining hypo-fractionated RT with ICI leads to a differential response in quantitative MRI (qMRI) parameters in HNC patients. These results suggest that qMRI parameter changes ten days following the start of RT may reflect a critical juncture in the anti-tumoral immune response when ICI is combined with hypo-fractionated RT.

MGACA-Net: a novel deep learning based multi-scale guided attention and context aggregation for localization of knee anterior cruciate ligament tears region in MRI images

Anterior cruciate ligament (ACL) tears are a common knee injury that can have serious consequences and require medical intervention. Magnetic resonance imaging (MRI) is the preferred method for ACL tear diagnosis. However, manual segmentation of the ACL in MRI images is prone to human error and can be time-consuming. This study presents a new approach that uses deep learning technique for localizing the ACL tear region in MRI images. The proposed multi-scale guided attention-based context aggregation (MGACA) method applies attention mechanisms at different scales within the DeepLabv3+ architecture to aggregate context information and achieve enhanced localization results. The model was trained and evaluated on a dataset of 917 knee MRI images, resulting in 15265 slices, obtaining state-of-the-art results with accuracy scores of 98.63%, intersection over union (IOU) scores of 95.39%, Dice coefficient scores (DCS) of 97.64%, recall scores of 97.5%, precision scores of 98.21%, and F1 Scores of 97.86% on validation set data. Moreover, our method performed well in terms of loss values, with binary cross entropy combined with Dice loss (BCE_Dice_loss) and Dice_loss values of 0.0564 and 0.0236, respectively, on the validation set. The findings suggest that MGACA provides an accurate and efficient solution for automating the localization of ACL in knee MRI images, surpassing other state-of-the-art models in terms of accuracy and loss values. However, in order to improve robustness of the approach and assess its performance on larger data sets, further research is needed.

Machine Learning-Based Performance Comparison to Diagnose Anterior Cruciate Ligament Tears

In recent times, knee joint pains have become severe enough to make daily tasks difficult. Knee osteoarthritis is a type of arthritis and a leading cause of disability worldwide. The middle of the knee contains a vital portion, the anterior cruciate ligament (ACL). It is necessary to diagnose the ACL ruptured tears early to avoid surgery. The study aimed to perform a comparative analysis of machine learning models to identify the condition of three ACL tears. In contrast to previous studies, this study also considers imbalanced data distributions as machine learning techniques struggle to deal with this problem. The paper applied and analyzed four machine learning classification models, namely, random forest (RF), categorical boosting (Cat Boost), light gradient boosting machines (LGBM), and highly randomized classifier (ETC) on the balanced, structured dataset of ACL. After oversampling a hyperparameter adjustment, the above four models have achieved an average accuracy of 95.72%, 94.98%, 94.98%, and 98.26%. There are 2070 observations and eight features in the collection of three diagnosis ACL classes after oversampling. The area under curve value was approximately 0.998, respectively. Experiments were performed using twelve machine learning algorithms with imbalanced and balanced datasets. However, the accuracy of the imbalanced dataset has remained under 76% for all twelve models. After oversampling, the proposed model may contribute to the investigation of ACL tears on magnetic resonance imaging and other knee ligaments efficiently and automatically without involving radiologists.

Software Measurement by Using Artificial Intelligence

Artificial intelligence (AI) is a subfield of computer science concerned with developing intelligent machines capable of performing tasks similar to those performed by humans. This human-created intelligence began more than 60 years ago. The goal of previous generations of applications was to demonstrate generic human-like behaviour. The goal has expanded with the advancement and increased compliance of this technology. It includes areas such as healthcare, gaming, and smart devices. The COVID-19 epidemic has posed a significant barrier to maintaining a sustainable strategy for mental health support clients with major mental illnesses and clinicians who have had to shift delivery modes quickly. In this study, we have conducted a systematic literature review (SLR) to provide an overview of the current state of the literature related to software measurement of healthcare using artificial intelligence. The study followed a secondary research strategy. The systematic literature review aim was to analyze software measurement of mental health illness in terms of previous literature. This study screened out of 28 research papers out of 1076 initial searches. We used Science Direct, IEEE Xplore, Springer Link, ACM, and Hindawi as database search engines. The research objective was to explore the needs of software applications and automation in the healthcare sector to bring efficiency to the systems. The research concluded that the healthcare setting crucially requires the implementation of software automation.

Automated Knee MR Images Segmentation of Anterior Cruciate Ligament Tears

The anterior cruciate ligament (ACL) is one of the main stabilizer parts of the knee. ACL injury leads to causes of osteoarthritis risk. ACL rupture is common in the young athletic population. Accurate segmentation at an early stage can improve the analysis and classification of anterior cruciate ligaments tears. This study automatically segmented the anterior cruciate ligament (ACL) tears from magnetic resonance imaging through deep learning. The knee mask was generated on the original Magnetic Resonance (MR) images to apply a semantic segmentation technique with convolutional neural network architecture U-Net. The proposed segmentation method was measured by accuracy, intersection over union (IoU), dice similarity coefficient (DSC), precision, recall and F1-score of 98.4%, 99.0%, 99.4%, 99.6%, 99.6% and 99.6% on 11451 training images, whereas on the validation images of 3817 was, respectively, 97.7%, 93.8%,96.8%, 96.5%, 97.3% and 96.9%. We also provide dice loss of training and test datasets that have remained 0.005 and 0.031, respectively. The experimental results show that the ACL segmentation on JPEG MRI images with U-Nets achieves accuracy that outperforms the human segmentation. The strategy has promising potential applications in medical image analytics for the segmentation of knee ACL tears for MR images.

Improved Deep Convolutional Neural Network to Classify Osteoarthritis from Anterior Cruciate Ligament Tear Using Magnetic Resonance Imaging

Anterior cruciate ligament (ACL) tear is caused by partially or completely torn ACL ligament in the knee, especially in sportsmen. There is a need to classify the ACL tear before it fully ruptures to avoid osteoarthritis. This research aims to identify ACL tears automatically and efficiently with a deep learning approach. A dataset was gathered, consisting of 917 knee magnetic resonance images (MRI) from Clinical Hospital Centre Rijeka, Croatia. The dataset we used consists of three classes: non-injured, partial tears, and fully ruptured knee MRI. The study compares and evaluates two variants of convolutional neural networks (CNN). We first tested the standard CNN model of five layers and then a customized CNN model of eleven layers. Eight different hyper-parameters were adjusted and tested on both variants. Our customized CNN model showed good results after a 25% random split using RMSprop and a learning rate of 0.001. The average evaluations are measured by accuracy, precision, sensitivity, specificity, and F1-score in the case of the standard CNN using the Adam optimizer with a learning rate of 0.001, i.e., 96.3%, 95%, 96%, 96.9%, and 95.6%, respectively. In the case of the customized CNN model, using the same evaluation measures, the model performed at 98.6%, 98%, 98%, 98.5%, and 98%, respectively, using an RMSprop optimizer with a learning rate of 0.001. Moreover, we also present our results on the receiver operating curve and area under the curve (ROC AUC). The customized CNN model with the Adam optimizer and a learning rate of 0.001 achieved 0.99 over three classes was highest among all. The model showed good results overall, and in the future, we can improve it to apply other CNN architectures to detect and segment other ligament parts like meniscus and cartilages.

Detection of COVID-19 in Chest X-ray Images: A Big Data Enabled Deep Learning Approach

Coronavirus disease (COVID-19) spreads from one person to another rapidly. A recently discovered coronavirus causes it. COVID-19 has proven to be challenging to detect and cure at an early stage all over the world. Patients showing symptoms of COVID-19 are resulting in hospitals becoming overcrowded, which is becoming a significant challenge. Deep learning's contribution to big data medical research has been enormously beneficial, offering new avenues and possibilities for illness diagnosis techniques. To counteract the COVID-19 outbreak, researchers must create a classifier distinguishing between positive and negative corona-positive X-ray pictures. In this paper, the Apache Spark system has been utilized as an extensive data framework and applied a Deep Transfer Learning (DTL) method using Convolutional Neural Network (CNN) three architectures -InceptionV3, ResNet50, and VGG19-on COVID-19 chest X-ray images. The three models are evaluated in two classes, COVID-19 and normal X-ray images, with 100 percent accuracy. But in COVID/Normal/pneumonia, detection accuracy was 97 percent for the inceptionV3 model, 98.55 percent for the ResNet50 Model, and 98.55 percent for the VGG19 model, respectively.

Efficient Detection of Knee Anterior Cruciate Ligament from Magnetic Resonance Imaging Using Deep Learning Approach

The most commonly injured ligament in the human body is an anterior cruciate ligament (ACL). ACL injury is standard among the football, basketball and soccer players. The study aims to detect anterior cruciate ligament injury in an early stage via efficient and thorough automatic magnetic resonance imaging without involving radiologists, through a deep learning method. The proposed approach in this paper used a customized 14 layers ResNet-14 architecture of convolutional neural network (CNN) with six different directions by using class balancing and data augmentation. The performance was evaluated using accuracy, sensitivity, specificity, precision and F1 score of our customized ResNet-14 deep learning architecture with hybrid class balancing and real-time data augmentation after 5-fold cross-validation, with results of 0.920%, 0.916%, 0.946%, 0.916% and 0.923%, respectively. For our proposed ResNet-14 CNN the average area under curves (AUCs) for healthy tear, partial tear and fully ruptured tear had results of 0.980%, 0.970%, and 0.999%, respectively. The proposing diagnostic results indicated that our model could be used to detect automatically and evaluate ACL injuries in athletes using the proposed deep-learning approach.

Final results of a multi-institutional phase II trial of reirradiation with concurrent weekly cisplatin and cetuximab for recurrent or second primary squamous cell carcinoma of the head and neck

Background

The optimal regimen of chemotherapy and reirradiation (re-XRT) for recurrent head and neck squamous cell carcinoma (HNSCC) is controversial. We report the final outcomes of a multicenter phase II trial evaluating cetuximab and cisplatin-based chemotherapy concurrent with re-XRT for patients with recurrent HNSCC.

Materials and methods

Patients with unresectable recurrent disease or positive margins after salvage surgery arising within a previously irradiated field with KPS ≥ 70 were eligible for this trial. Cetuximab 400 mg/m2 was delivered as a loading dose in week 1 followed by weekly cetuximab 250 mg/m2 and cisplatin 30 mg/m2 concurrent with 6 weeks of intensity-modulated radiotherapy to a dose of 60-66 Gy in 30 daily fractions. Patients who previously received both concurrent cetuximab and cisplatin with radiation or who received radiotherapy less than 6 months prior were ineligible.

Results

From 2009 to 2013, 48 patients enrolled on this trial, 2 did not receive any protocol treatment. Of the remaining 46 patients, 34 were male and 12 female, with a median age of 62 years (range 36-85). Treatment was feasible and only 1 patient did not complete the treatment course. Common grade 3 or higher acute toxicities were lymphopenia (46%), pain (22%), dysphagia (13%), radiation dermatitis (13%), mucositis (11%) and anorexia (11%). There were no grade 5 acute toxicities. Eight grade 3 late toxicities were observed, four of which were swallowing related. With a median follow-up of 1.38 years, the 1-year overall survival (OS) was 60.4% and 1-year recurrence-free survival was 34.1%. On univariate analysis, OS was significantly improved with young age (P = 0.01). OS was not associated with radiation dose, surgery before re-XRT or interval from prior XRT.

Conclusions

Concurrent cisplatin and cetuximab with re-XRT is feasible and offers good treatment outcomes for patients with high-risk features. Younger patients had significantly improved OS.

ClinicalTrials.Gov Identifier

NCT00833261.

Health status and renal function evaluation of kidney vendors: a report from Pakistan

Unrelated kidney transplants have lead to commerce and kidney vending in Pakistan. This study on 104 vendors reports demographics, history, physical and systemic examination, ultrasound findings, renal and liver function and GFR by Cockcroft-Gault. Results were compared with 184 age, sex and nephrectomy duration matched living-related donors controls. Comparison of vendors versus controls showed mean age of 30.55 +/- 8.1 versus 30.65 +/- 7.85 (p = 0.91) years, M:F of 4.5:1 versus 4.2:1 and nephrectomy period of 33.89 +/- 30 versus 32.01 +/- 29.71 (p = 0.60) months respectively. Of the vendors 67% were bonded laborers earning <50 $/month as compared to controls where 68% were skilled laborers and self-employed earning >100 $/month. History of vendors revealed jaundice in 8%, stone disease in 2% and urinary tract symptoms in 4.8%. Postnephrectomy findings between vendors versus donors showed BMI of 21.02 +/- 2.8 versus 23.02 +/- 4.2 (p = 0.0001), hypertension in 17% versus 9.2% (p = 0.04), serum creatinine (mg/dL) of 1.17+/-0.21 versus 1.02 +/- 0.27 (p = 0.0001), GFR (mL/min) of 70.94 +/- 14.2 versus 95.4 +/- 20.44 (p = 0.0001), urine protein/creatinine of 0.150 +/- 0.109 versus 0.10 +/- 0.10 (p = 0.0001), hepatitis C positivity in 27% versus 1.0% (p = 0.0001) and hepatitis B positive 5.7% versus 0.5% (p = 0.04), respectively. In conclusion, vendors had compromised renal function suggesting inferior selection and high risk for developing chronic kidney disease in long term.