Creation of synthetic contrast-enhanced computed tomography images using deep neural networks to screen for renal cell carcinoma.

In this study, we elucidate if synthetic contrast enhanced computed tomography images created from plain computed tomography images using deep neural networks could be used for screening, clinical diagnosis, and postoperative follow-up of small-diameter renal tumors. This retrospective, multicenter study included 155 patients (artificial intelligence training cohort [n = 99], validation cohort [n = 56]) who underwent surgery for small-diameter (≤40 mm) renal tumors, with the pathological diagnosis of renal cell carcinoma, during 2010-2020. We created a learned deep neural networks using pix2pix.

Low-invasive neural recording in mouse models with diabetes via an ultrasmall needle-electrode.

Diabetes is known to cause a variety of complications, having a high correlation with Alzheimer's disease. Electrophysiological recording using a microscale needle electrode is a promising technology for the study, however, diabetic brain tissue is more difficult to record neuronal activities than normal tissue due to these complications including the development of cerebrovascular disease. Here we show an electrophysiological methodology for diabetic db/db mice (+Lepr/+Lepr) using a 4-μm-tip diameter needle-electrode device.

Heuristic Method for Minimizing Model Size of CNN by Combining Multiple Pruning Techniques.

Network pruning techniques have been widely used for compressing computational and memory intensive deep learning models through removing redundant components of the model. According to the pruning granularity, network pruning can be categorized into structured and unstructured methods. The structured pruning removes the large components in a model such as channels or layers, which might reduce the accuracy.

Nanoscale-tipped wire array injections transfer DNA directly into brain cells ex vivo and in vivo.

Genetic modification to restore cell functions in the brain can be performed through the delivery of biomolecules in a minimally invasive manner into live neuronal cells within brain tissues. However, conventional nanoscale needles are too short (lengths of ~10 µm) to target neuronal cells in ~1-mm-thick brain tissues because the neuronal cells are located deep within the tissue. Here, we report the use of nanoscale-tipped wire (NTW) arrays with diameters < 100 nm and wire lengths of ~200 µm to address biomolecule delivery issues.

A floating 5 μm-diameter needle electrode on the tissue for damage-reduced chronic neuronal recording in mice.

Microelectrode technology is essential in electrophysiology and has made contributions to neuroscience as well as to medical applications. However, it is necessary to minimize tissue damage associated with needle-like electrode on the brain tissue and the implantation surgery, which makes stable chronic recording impossible. Here, we report on an approach for using a 5 μm-diameter needle electrode, which enables the following of tissue motions, a surgical method.