A screening assistance system for cervical cytology of squamous cell atypia based on a two-step combined CNN algorithm with label smoothing.

BACKGROUND: Although many cervical cytology diagnostic support systems have been developed, it is challenging to classify overlapping cell clusters with a variety of patterns in the same way that humans do. In this study, we developed a fast and accurate system for the detection and classification of atypical cell clusters by using a two-step algorithm based on two different deep learning algorithms. METHODS: We created 919 cell images from liquid-based cervical cytological samples collected at Sapporo Medical University and annotated them based on the Bethesda system as a dataset for machine learning. Most of the images captured overlapping and crowded cells, and images were oversampled by digital processing. The detection system consists of two steps: (1) detection of atypical cells using You Only Look Once v4 (YOLOv4) and (2) classification of the detected cells using ResNeSt. A label smoothing algorithm was used for the dataset in the second classification step. This method annotates multiple correct classes from a single cell image with a smooth probability distribution. RESULTS: The first step, cell detection by YOLOv4, was able to detect all atypical cells above ASC-US without any observed false negatives. The detected cell images were then analyzed in the second step, cell classification by the ResNeSt algorithm, which exhibited average accuracy and F-measure values of 90.5% and 70.5%, respectively. The oversampling of the training image and label smoothing algorithm contributed to the improvement of the system's accuracy. CONCLUSION: This system combines two deep learning algorithms to enable accurate detection and classification of cell clusters based on the Bethesda system, which has been difficult to achieve in the past. We will conduct further research and development of this system as a platform for augmented reality microscopes for cytological diagnosis.

Integrated datasets of normalized brain with functional localization using intra-operative electrical stimulation.

PURPOSE: The purpose of this study was to transform brain mapping data into a digitized intra-operative MRI and integrated brain function dataset for predictive glioma surgery considering tumor resection volume, as well as the intra-operative and postoperative complication rates. METHODS: Brain function data were transformed into digitized localizations on a normalized brain using a modified electric stimulus probe after brain mapping. This normalized brain image with functional information was then projected onto individual patient's brain images including predictive brain function data. RESULTS: Log data were successfully acquired using a medical device integrated into intra-operative MR images, and digitized brain function was converted to a normalized brain data format in 13 cases. For the electrical stimulation positions in which patients showed speech arrest (SA), speech impairment (SI), motor and sensory responses during cortical mapping processes in awake craniotomy, the data were tagged, and the testing task and electric current for the stimulus were recorded. There were 13 SA, 7 SI, 8 motor and 4 sensory responses (32 responses) in total. After evaluation of transformation accuracy in 3 subjects, the first transformation from intra- to pre-operative MRI using non-rigid registration was calculated as 2.6 ± 1.5 and 2.1 ± 0.9 mm, examining neighboring sulci on the electro-stimulator position and the cortex surface near each tumor, respectively; the second transformation from pre-operative to normalized brain was 1.7 ± 0.8 and 1.4 ± 0.5 mm, respectively, representing acceptable accuracy. CONCLUSION: This image integration and transformation method for brain normalization should facilitate practical intra-operative brain mapping. In the future, this method may be helpful for pre-operatively or intra-operatively predicting brain function.

Positioning error evaluation of GPU-based 3D ultrasound surgical navigation system for moving targets by using optical tracking system.

PURPOSE: A near real-time three-dimensional (3D) ultrasound navigation system has been developed for guiding surgery involving internal organs that move and change shape (e.g., abdominal surgery, fetal surgery). In practical applications, significant errors arise between the actual navigation-image positions depending on the time delay of the system. Therefore, the positioning error of the system relative to the target velocity was evaluated. METHODS: We developed a method for evaluating the positioning error of a graphics processing unit-based 3D ultrasound surgical navigation system (with an optical tracking system) for moving targets. The effectiveness of this system was quantitatively evaluated in terms of its image processing runtime, target registration error (TRE), and positioning error for a moving target. The positioning error was evaluated for a phantom (with an optical tracking marker) moving at speeds of 5-25 mm/s, and the navigation target was the center point of the phantom. The imaging range of the volume data was set to the maximum angle and range of the ultrasound diagnostic system (update rate: 4 Hz). RESULTS: The image processing runtime was 27.43 ± 4.80 ms, and the TRE was 1.50 ± 0.28 mm. The positioning error was 4.24 ± 0.12 mm for a target moving at a speed of 10 mm/s and 5.36 ± 0.10 mm for one moving at 15 mm/s. CONCLUSION: The effectiveness of an ultrasound navigation system was quantitatively evaluated by using the positioning error for a moving target. This navigation system demonstrated high calculation speed and positioning accuracy for a moving target. Therefore, it is suitable to guide the surgery of abdominal internal organs (e.g., in fetal and abdominal surgeries) that move or change shape during breathing and surgical approaches.

GPU based real-time surgical navigation system with three-dimensional ultrasound imaging for water-filled laparo-endoscope surgery.

Presently, a variety of navigation systems are employed in clinical treatments involving neurosurgery, ENT, orthopedic, and head and neck surgery. An ultrasound diagnostic system is used as the navigation system for movable and deformable organs in the abdomen or chest. In this study, we developed a real-time updated 3D ultrasound navigation system that facilitates the high-speed transfer of image data and GPGPU processing for fetal surgery and water-filled laparo-endoscopic surgery (WAFLES). Experimental results showed that our system was able to update every 62 ms. Further, in vivo experimental results showed the ability of our system to guide a surgeon to a target organ during WAFLES in a case where the endoscopic view experienced problems.

Assessing the roles of presynaptic ryanodine receptors and adenosine receptors in caffeine-induced enhancement of hippocampal mossy fiber transmission.

Caffeine robustly enhances transmitter release from the hippocampal mossy fiber terminals, although it remains uncertain whether calcium mobilization through presynaptic ryanodine receptors mediates this enhancement. In this study, we adopted a selective adenosine A1 blocker to assess relative contribution of A1 receptors and ryanodine receptors in caffeine-induced synaptic enhancement. Application of caffeine further enhanced transmission at the hippocampal mossy fiber synapse even after full blockade of adenosine A1 receptors. This result suggests that caffeine enhances mossy fiber synaptic transmission by two distinct presynaptic mechanisms, i.e., removal of A1 receptor-mediated tonic inhibition and ryanodine receptor-mediated calcium release from intracellular stores.