Objective: To compare the performance of the deep learning-based lesion detection algorithm (DLLD) in detecting liver metastasis with that of radiologists. Materials and Methods: This clinical retrospective study used 4386-slice computed tomography (CT) images and labels from a training cohort (502 patients with colorectal cancer [CRC] from November 2005 to December 2010) to train the DLLD for detecting liver metastasis, and used CT images of a validation cohort (40 patients with 99 liver metastatic lesions and 45 patients without liver metastasis from January 2011 to December 2011) for comparing the performance of the DLLD with that of readers (three abdominal radiologists and three radiology residents). For per-lesion binary classification, the sensitivity and false positives per patient were measured. Results: A total of 85 patients with CRC were included in the validation cohort. In the comparison based on per-lesion binary classification, the sensitivity of DLLD (81.82%, [81/99]) was comparable to that of abdominal radiologists (80.81%, p = 0.80) and radiology residents (79.46%, p = 0.57). However, the false positives per patient with DLLD (1.330) was higher than that of abdominal radiologists (0.357, p < 0.001) and radiology residents (0.667, p < 0.001). Conclusion DLLD showed a sensitivity comparable to that of radiologists when detecting liver metastasis in patients initially diagnosed with CRC. However, the false positives of DLLD were higher than those of radiologists. Therefore, DLLD could serve as an assistant tool for detecting liver metastasis instead of a standalone diagnostic tool.

Objective: To compare the performance of the deep learning-based lesion detection algorithm (DLLD) in detecting liver metastasis with that of radiologists. Materials and Methods: This clinical retrospective study used 4386-slice computed tomography (CT) images and labels from a training cohort (502 patients with colorectal cancer [CRC] from November 2005 to December 2010) to train the DLLD for detecting liver metastasis, and used CT images of a validation cohort (40 patients with 99 liver metastatic lesions and 45 patients without liver metastasis from January 2011 to December 2011) for comparing the performance of the DLLD with that of readers (three abdominal radiologists and three radiology residents). For per-lesion binary classification, the sensitivity and false positives per patient were measured. Results: A total of 85 patients with CRC were included in the validation cohort. In the comparison based on per-lesion binary classification, the sensitivity of DLLD (81.82%, [81/99]) was comparable to that of abdominal radiologists (80.81%, p = 0.80) and radiology residents (79.46%, p = 0.57). However, the false positives per patient with DLLD (1.330) was higher than that of abdominal radiologists (0.357, p < 0.001) and radiology residents (0.667, p < 0.001). Conclusion DLLD showed a sensitivity comparable to that of radiologists when detecting liver metastasis in patients initially diagnosed with CRC. However, the false positives of DLLD were higher than those of radiologists. Therefore, DLLD could serve as an assistant tool for detecting liver metastasis instead of a standalone diagnostic tool.

BACKGROUND: Diabetic neuropathy (DN) is the most common complication of diabetes, and approximately 50% of patients with this disease suffer from peripheral neuropathy. Nerve fiber loss in DN occurs due to myelin defects and is characterized by symptoms of impaired nerve function. Schwann cells (SCs) are the main support cells of the peripheral nervous system and play important roles in several pathways contributing to the pathogenesis and development of DN. We previously reported that human tonsil-derived mesenchymal stem cells differentiated into SCs (TMSC-SCs), named neuronal regeneration-promoting cells (NRPCs), which cells promoted nerve regeneration in animal models with peripheral nerve injury or hereditary peripheral neuropathy. METHODS: In this study, NRPCs were injected into the thigh muscles of BKS-db/db mice, a commonly used type 2 diabetes model, and monitored for 26 weeks. Von Frey test, sensory nerve conduction study, and staining of sural nerve, hind foot pad, dorsal root ganglia (DRG) were performed after NRPCs treatment. RESULTS: Von Frey test results showed that the NRPC treatment group (NRPC group) showed faster responses to less force than the vehicle group. Additionally, remyelination of sural nerve fibers also increased in the NRPC group. After NRPCs treatment, an improvement in response to external stimuli and pain sensation was expected through increased expression of PGP9.5 in the sole and TRPV1 in the DRG. CONCLUSION The NRPCs treatment may alleviate DN through the remyelination and the recovery of sensory neurons, could provide a better life for patients suffering from complications of this disease.

BACKGROUND: Retinal degenerative disease (RDD), one of the most common causes of blindness, is predominantly caused by the gradual death of retinal pigment epithelial cells (RPEs) and photoreceptors due to various causes. Cell-based therapies, such as stem cell implantation, have been developed for the treatment of RDD, but potential risks, including teratogenicity and immune reactions, have hampered their clinical application. Stem cell-derived extracellular vesicles (EVs) have recently emerged as a cell-free alternative therapeutic strategy; however, additional invasiveness and low yield of the stem cell extraction process is problematic. METHODS: To overcome these limitations, we developed therapeutic EVs for the treatment of RDD which were extracted from tonsil-derived mesenchymal stem cells obtained from human tonsil tissue discarded as medical waste following tonsillectomy (T-MSC EVs). To verify the biocompatibility and cytoprotective effect of T-MSC EVs, we measured cell viability by co-culture with human RPE without or with toxic all-trans-retinal. To elucidate the cytoprotective mechanism of T-MSC EVs, we performed transcriptome sequencing using RNA extracted from RPEs. The in vivo protective effect of T-MSC EVs was evaluated using Pde6b gene knockout rats as an animal model of retinitis pigmentosa. RESULTS: T-MSC EVs showed high biocompatibility and the human pigment epithelial cells were significantly protected in the presence of T-MSC EVs from the toxic effect of all-trans-retinal. In addition, T-MSC EVs showed a dosedependent cell death-delaying effect in real-time quantification of cell death. Transcriptome sequencing analysis revealed that the efficient ability of T-MSC EVs to regulate intracellular oxidative stress may be one of the reasons explaining their excellent cytoprotective effect. Additionally, intravitreally injected T-MSC EVs had an inhibitory effect on the destruction of the outer nuclear layer in the Pde6b gene knockout rat. CONCLUSIONS Together, the results of this study indicate the preventive and therapeutic effects of T-MSC EVs during the initiation and development of retinal degeneration, which may be a beneficial alternative for the treatment of RDD.