Purpose. In this study, we proposed an automated deep learning (DL) method for head and neck cancer (HNC) gross tumor volume (GTV) contouring on positron emission tomography-computed tomography (PET-CT) images. Materials and Methods. PET-CT images were collected from 22 newly diagnosed HNC patients, of whom 17 (Database 1) and 5 (Database 2) were from two centers, respectively. An oncologist and a radiologist decided the gold standard of GTV manually by consensus. We developed a deep convolutional neural network (DCNN) and trained the network based on the two-dimensional PET-CT images and the gold standard of GTV in the training dataset. We did two experiments: Experiment 1, with Database 1 only, and Experiment 2, with both Databases 1 and 2. In both Experiment 1 and Experiment 2, we evaluated the proposed method using a leave-one-out cross-validation strategy. We compared the median results in Experiment 2 (GTVa) with the performance of other methods in the literature and with the gold standard (GTVm). Results. A tumor segmentation task for a patient on coregistered PET-CT images took less than one minute. The dice similarity coefficient (DSC) of the proposed method in Experiment 1 and Experiment 2 was 0.481∼0.872 and 0.482∼0.868, respectively. The DSC of GTVa was better than that in previous studies. A high correlation was found between GTVa and GTVm (R = 0.99, P<0.001). The median volume difference (%) between GTVm and GTVa was 10.9%. The median values of DSC, sensitivity, and precision of GTVa were 0.785, 0.764, and 0.789, respectively. Conclusion. A fully automatic GTV contouring method for HNC based on DCNN and PET-CT from dual centers has been successfully proposed with high accuracy and efficiency. Our proposed method is of help to the clinicians in HNC management.

Objectives. To evaluate the application of a deep learning architecture, based on the convolutional neural network (CNN) technique, to perform automatic tumor segmentation of magnetic resonance imaging (MRI) for nasopharyngeal carcinoma (NPC). Materials and Methods. In this prospective study, 87 MRI containing tumor regions were acquired from newly diagnosed NPC patients. These 87 MRI were augmented to >60,000 images. The proposed CNN network is composed of two phases: feature representation and scores map reconstruction. We designed a stepwise scheme to train our CNN network. To evaluate the performance of our method, we used case-by-case leave-one-out cross-validation (LOOCV). The ground truth of tumor contouring was acquired by the consensus of two experienced radiologists. Results. The mean values of dice similarity coefficient, percent match, and their corresponding ratio with our method were 0.89±0.05, 0.90±0.04, and 0.84±0.06, respectively, all of which were better than reported values in the similar studies. Conclusions. We successfully established a segmentation method for NPC based on deep learning in contrast-enhanced magnetic resonance imaging. Further clinical trials with dedicated algorithms are warranted.

In recent years, diagnostic and therapeutic approaches for rheumatoid arthritis (RA) have continued to improve. However, in the advanced stages of the disease, patients are unable to achieve long-term clinical remission and often suffer from systemic multi-organ damage and severe complications. Patients with RA usually have no overt clinical manifestations in the early stages, and by the time a definitive diagnosis is made, the disease is already at an advanced stage. RA is diagnosed clinically and with laboratory tests, including the blood markers C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) and the autoantibodies rheumatoid factor (RF) and anticitrullinated protein antibodies (ACPA). However, the presence of RF and ACPA autoantibodies is associated with aggravated disease, joint damage, and increased mortality, and these autoantibodies have low specificity and sensitivity. The etiology of RA is unknown, with the pathogenesis involving multiple factors and clinical heterogeneity. The early diagnosis, subtype classification, and prognosis of RA remain challenging, and studies to develop minimally invasive or non-invasive biomarkers in the form of biofluid biopsies are becoming more common. Non-coding RNA (ncRNA) molecules are composed of long non-coding RNAs, small nucleolar RNAs, microRNAs, and circular RNAs, which play an essential role in disease onset and progression and can be used in the early diagnosis and prognosis of RA. In this review of the diagnostic and prognostic approaches to RA disease, we provide an overview of the current knowledge on the subject, focusing on recent advances in mRNA–ncRNA as diagnostic and prognostic biomarkers from the biofluid to the tissue level.

Deep vein thrombosis (DVT) is a common but elusive illness that can result in long-term disability or death. Accurate detection of thrombosis and assessment of its size and distribution are critical for treatment decision-making. In the present study, we sought to develop and evaluate a cardiovascular magnetic resonance (CMR) black-blood thrombus imaging (BTI) technique, based on delay alternating with nutation for tailored excitation black-blood preparation and variable flip angle turbo-spin-echo readout, for the diagnosis of non-acute DVT. This prospective study was approved by institutional review board and informed consent obtained from all subjects. BTI was first conducted in 11 healthy subjects for parameter optimization and then conducted in 18 non-acute DVT patients to evaluate its diagnostic performance. Two clinically used CMR techniques, contrast-enhanced CMR venography (CE-MRV) and three dimensional magnetization prepared rapid acquisition gradient echo (MPRAGE), were also conducted in all patients for comparison. All images obtained from patients were analyzed on a per-segment basis. Using the consensus diagnosis of CE-MRV as the reference, the sensitivity (SE), specificity (SP), positive and negative predictive values (PPV and NPV), and accuracy (ACC) of BTI and MPRAGE as well as their diagnostic agreement with CE-MRV were calculated. Besides, diagnostic confidence and interreader diagnostic agreement were evaluated for all three techniques. BTI with optimized parameters effectively nulled the venous blood flow signal and allowed directly visualizing the thrombus within the black-blood lumen. Higher SE (90.4% vs 67.6%), SP (99.0% vs. 97.4%), PPV (95.4% vs. 85.6%), NPV (97.8% vs 92.9%) and ACC (97.4% vs. 91.8%) were obtained by BTI in comparison with MPRAGE. Good diagnostic confidence and excellent diagnostic and interreader agreements were achieved by BTI, which were superior to MPRAGE on detecting the chronic thrombus. BTI allows direct visualization of non-acute DVT within the dark venous lumen and has the potential to be a reliable diagnostic tool without the use of contrast medium.

Purpose: To evaluate the image resolutions in global teleophthalmology using 2 smartphones. Methods: A standard resolution test card and two human eyes (one eye wearing a contact lens) were imaged with an iPhone 4S adapted for slitlamp biomicroscopy in Hangzhou, Zhejiang, China. An iPhone 5 was used as a receptor on the other side of the world (Miami, FL). The real-time images and still images were transferred through Skype from Hangzhou to Miami during slitlamp eye examination. The real-time resolutions captured and displayed on the both phones at different places were measured. There was vocal communication concurrent between two parties during image transference and the examination by the slitlamp biomicroscopy performed in Hangzhou. Results: The real-time and still images were transferred between two cities while two operators talked with each other smoothly and without difficulty. The viewer in Miami was able to instruct the operator in Hangzhou to thoroughly examine the eye using the slitlamp microscopy with different magnification settings and illumination settings. The resolution of the still images recorded in the iPhone in Hangzhou was higher than that of the real-time images on the screen. The main features of the eye were recognizable in real-time. Conclusions: The study demonstrated that global teleophthalmology is feasible using two smartphones. The system is simple, portable, and affordable, and the image quality in still and real-time images is acceptable for real-time teleophthalmology.