Early and accurate diagnosis of osteosarcomas (OS) is of great clinical significance, and machine learning (ML) based methods are increasingly adopted. However, current ML-based methods for osteosarcoma diagnosis consider only X-ray images, usually fail to generalize to new cases, and lack explainability. In this paper, we seek to explore the capability of deep learning models in diagnosing primary OS, with higher accuracy, explainability, and generality. Concretely, we analyze the added value of integrating the biochemical data, i.e., alkaline phosphatase (ALP) and lactate dehydrogenase (LDH), and design a model that incorporates the numerical features of ALP and LDH and the visual features of X-ray imaging through a late fusion approach in the feature space. We evaluate this model on real-world clinic data with 848 patients aged from 4 to 81. The experimental results reveal the effectiveness of incorporating ALP and LDH simultaneously in a late fusion approach, with the accuracy of the considered 2608 cases increased to 97.17%, compared to 94.35% in the baseline. Grad-CAM visualizations consistent with orthopedic specialists further justified the model's explainability.

This study aimed to evaluate the performance of 18F-MD-PSMA PET/CT in patients previously treated for prostate cancer by either surgery or therapy, but later relapsed biochemically. This retrospective study enrolled 213 patients in sequence previously treated for prostate cancer by either surgery or therapy, but later PSA relapsed. A total of 191 of these 213 patients were included in this analysis. All patients were biochemically relapsed after radical prostatectomy or therapy, had 18F-MD-PSMA PET/CT scan within 1 week, and were off hormonal therapy at the time of the scans. The new tracer was compared directly with 11C-choline in sensitivity. In 3 patients, a side-by-side comparison between 18F-MD-PSMA and 11C-choline was performed, and it was found that the former was about 3 times more sensitive than the latter. The analysis of PET imaging using 18F-MD-PSMA in 191 relapsed patients showed that less than 10% of patients showed the disease limited in the prostate. Among the remote lesions, the number in decreasing order was bone, followed by lymph nodes and other organs. The maximal SUV in lesions in each patient followed an exponential decay, with SUV inclined to the lower end. The Gleason score measured at the diagnosis showed no correlation with the average number of lesions in each patient, the average maximal SUV values among this cohort of patients, and the PSA values measured at the time of PET imaging. The number of lesions observed in each patient has no correlation with the PSA value measured at the time of PET imaging. When PSA value was measured as an independent biomarker at the time of PET imaging, the positivity of PET imaging using 18F-MD-PSMA increased along with an increase in PSA value, but with exceptions where PSMA expression was low or negative. From the PET imaging of this radioligand, the majority of patients showed oligo-metastasis, favoring using local therapy to manage the disease. An 18F-MD-PSMA as a radioligand was found to be superior to 11C-choline in the setting of patients with biochemical relapse after previous treatment. Its PET imaging results matched those of established PSMA radioligands, but its chemical structure was found to have added features to conjugate with other functional molecules, such as those with therapeutic properties. This radioligand lays the foundation for our further work.

Astrocytes derive from different lineages and play a critical role in neuropathic pain after spinal cord injury (SCI). Whether selectively eliminating these main origins of astrocytes in lumbar enlargement could attenuate SCI-induced neuropathic pain remains unclear. Through transgenic mice injected with an adeno-associated virus vector and diphtheria toxin, astrocytes in lumbar enlargement were lineage traced, targeted, and selectively eliminated. Pain-related behaviors were measured with an electronic von Frey apparatus and a cold/hot plate after SCI. RNA sequencing, bioinformatics analysis, molecular experiment, and immunohistochemistry were used to explore the potential mechanisms after astrocyte elimination. Lineage tracing revealed that the resident astrocytes but not ependymal cells were the main origins of astrocytes-induced neuropathic pain. SCI-induced mice to obtain significant pain symptoms and astrocyte activation in lumbar enlargement. Selective resident astrocyte elimination in lumbar enlargement could attenuate neuropathic pain and activate microglia. Interestingly, the type I interferons (IFNs) signal was significantly activated after astrocytes elimination, and the most activated Gene Ontology terms and pathways were associated with the type I IFNs signal which was mainly activated in microglia and further verified in vitro and in vivo. Furthermore, different concentrations of interferon and Stimulator of interferon genes (STING) agonist could activate the type I IFNs signal in microglia. These results elucidate that selectively eliminating resident astrocytes attenuated neuropathic pain associated with type I IFNs signal activation in microglia. Targeting type I IFNs signals is proven to be an effective strategy for neuropathic pain treatment after SCI.

In this paper, we present a promising method for modulating the frequency of magnetic droplets using strain induced by an adjacent piezoelectric layer. This strain results in significant changes in droplet frequency (∼6 GHz) and variations in diameter or mode, while maintaining a fixed current density of 1.4×1012A/m2 under both high and low fields. By applying strain pulses, we successfully modulate the magnetic droplet frequency, paving the way for potential implementation of digital binary frequency shift-keying modulation technique with magnetic droplets. Moreover, the effects of Oersted field and the Zhang-Li torque induced by lateral current have also been discussed, showing a minimal impact on the oscillation frequencies, especially at high fields. Published by the American Physical Society 2024