PidPort is a pathology image management cloud system that can be used as an optimal repository for digital pathological specimens. Multiple healthcare professionals can view and share cloud-stored pathological specimens anytime and anywhere. It can be utilized for a wide range of purposes such as remote consultations, academic meetings, conferences, and educational lectures across facilities. An efficient and rapid pathological flow is fully supported.
By storing pathology image data and case information on the cloud, PidPort enables the secure management and rapid search of information. No need to install an application, just access PidPort on your PC or tablet. There is no need for initial costs such as equipment installation and you can use it anytime and anywhere in any environment where the Internet is available. We will reduce the workload of pathologists by creating a comfortable environment for pathological diagnoses at a low cost and promptly.
Introducing PidPort enables the easy sharing of diagnoses and cases anytime and anywhere. It comes standard with its own high-speed viewer with excellent visibility and operability, enabling smooth and comfortable pathological diagnoses on the cloud. Timely and smooth consultations between facilities is possible, leading to the resolution of problems that have previously required the transportation of pathological specimens and business trips to a remote location.
We have developed our own AI for a number of disorders that analyzes pathological image data immediately with high accuracy. Through collaborative research with multiple medical institutions, it creates digital images based on hundreds of thousands of pathological specimens. Furthermore, a large number of highly specialized pathologists create teacher data in order to develop pathological AI by deep learning/transfer learning of unique technology. We are willing to contribute to the development of AI in order to be a partner for pathologists so that it can eliminate their workloads and psychological burdens in the future.
By building a strong mutual support network with group hospitals and related hospitals around the world, we will make daily diagnostic work more comfortable and reduce the burden on pathologists in the field.
This allows you to create an environment where you can easily hold and participate in conferences from your homes and other remote locations.
Significant learning effects are expected by establishing an environment where you can learn and carefully observe all pathological diagnostic specimens on a virtual slide rather than learning by microscopic photography with a limited field of view.
Taking advantage of the characteristics of the cloud enables improvement in the efficiency and speed of collaborative research
It comes standard with our own high-speed viewer that is highly visible and user-friendly, with no need for soft installation, so you can use it comfortably.
Commenting on a case allows users to communicate with each other online. It can be used for a wide range of purposes such as consultations, academic meetings, educational research, etc.
A lot of drawing tools are available for you to graphically illustrate and write in images. Written annotations (notes) can be shared between users.
You can upload pathology images and case data all at once.
You can tag case folders and set favorites, so they will be searched and displayed all at once.
Documents related to cases can be attached. You can comprehensively view virtual slides and necessary materials.
You can adjust the color to your liking and save the color settings separately for both shared and private use.
It is possible to display up to four virtual slides at the same time. A synchronization mode is equipped that enables the operation of all images at the same time.
Screenshots make it easy to save the parts you want to keep.
A Deep Learning Model for Prostate Adenocarcinoma Classification in Needle Biopsy Whole-Slide Images Using Transfer Learning
The histopathological diagnosis of prostate adenocarcinoma in needle biopsy specimens is of pivotal importance for determining optimum prostate cancer treatment. Since diagnosing a large number of cases containing 12 core biopsy specimens by pathologists using a microscope is time-consuming manual system and limited in terms of human resources, it is necessary to develop new techniques that can rapidly and accurately screen large numbers of histopathological prostate needle biopsy specimens. Computational pathology applications that can assist pathologists in detecting and classifying prostate adenocarcinoma from whole-slide images (WSIs) would be of great benefit for routine pathological practice. In this paper, we trained deep learning models capable of classifying needle biopsy WSIs into adenocarcinoma and benign (non-neoplastic) lesions. We evaluated the models on needle biopsy, transurethral resection of the prostate (TUR-P), and The Cancer Genome Atlas (TCGA) public dataset test sets, achieving an ROC-AUC up to 0.978 in needle biopsy test sets and up to 0.9873 in TCGA test sets for adenocarcinoma.
Masayuki Tsuneki, Board of Director
He worked at a national university, Yale University, and National Cancer Center as a pathology researcher, a pathologist and an academic teacher. He had been working on research in molecular pathology, particularly related to carcinogenesis. He is eager to create a new technology that can provide objective diagnoses in the field of clinical pathology based on scientific evidence. With his professional knowledge of human pathology, his strength is in international academic research.