3DevS sp. z o. o.

SOFTWARE TECHNOLOGY

fitted to suit
Your needs

+48 605 230 010

+48 606 961 328

biuro@3devs.pl

Work stages

we present 6 basic stages of work on the application / software

Regardless of which service you choose from our offer, the various stages of work will have a similar course. Depending on the scale of the project, the type of application, the IT environment, etc., some steps may be omitted, others may be more extensive, and the final valuation may vary. Familiarize yourself with the course of our joint work.

more

1
PROBLEM

solution

The most important is to recognition of the problem. At this stage we talk with the client, we try to recognize the client's needs and what problems and obstacles he faces.

2
PROBLEM ANALYSIS

strengths, weaknesses, opportunities, threats

We carefully analyze what are the possible solutions for the problem, present the possibilities and together with the client we choose the most reliable ones for his needs.

3
DESIGNING

UI design - look and behaviour

Before we start programming we design the look and behavior of the application. The solution must meet expectations and be the best suited to the needs of the customer. Therefore, at this stage, it is necessary to consult the client before coding.

4
PROGRAMMING

.NET, .NET Core, C#, CSS3, HTML5, JS,

Our best specialists deal with the most labor-intensive stage of program creation. They select the most appropriate technologies and algorithms.

5
TESTING

perfecting the details

Before a customer receives a finished product from us, it is subject to precise and rigorous tests in terms of operation, stability and intuitiveness.

6
FINISHED PRODUCT

further development, training, updates

The customer gets a finished product - it can be a program, an application, a management system, a web site or a database. However, at this stage, our work does not end, because with the development of the client's company, it is possible to continue the product development.
MACHNE LEARNING

Case Study - Car parts storage system

Algorithms learn from the data they process, and when they make decisions they use what they've learned.

Business challenge

The client was looking for a solution to three problems:

  • the possibility of registering used car parts in the warehouse system
  • the method of transferring used car parts in warehouse processes
  • reducing the time-consuming and often ineffective process of manually describing each used car part

Our approach

We have made an innovative solution for recording photos of used car parts made with the camera and putting them in this form in the register of the warehouse. Using the algorithms ML (Machine Learning) we have prepared prompts for new images of used car parts based on previously used photos of used car parts.

Results

ML expands its algorithms and provides more and more accurate recommendations for newly taken photos. The solution used limits the time to describe the part by 30%.

3D

Case Study - Image navigation of cancerous changes

Business challenge

The challenge was to develop an image navigation system to help fight cancerous changes in the liver. The surgeon, thanks to the visualization of the model, would be able to select the optimal place of puncture/incisions at the planning stage. This would reduce the risks associated with surgery and reduce the risk of complications.

Our approach

For this purpose, we have developed a mechanism for generating a personalized anatomical model of the patient based on a tomographic examination. During the procedure, the system tracks the position of the tools relative to the patient's body and then visualizes it in relation to the anatomical model. This gives the surgeon feedback and "navigates" him during the surgery. The system also helps in the precise insertion of the ablation needle, which thanks to the flow of electricity heats the change to the temperature of protein coagulation, destroying cancer cells. The task was not trivial, because as a result of breathing movements organs move even by a few centimeters during each breathing cycle (lasting about 4s). Therefore, it was necessary to develop a mechanism determining the ideal puncture moment. In addition, the tomographic examination is performed in a fraction of a second (about 300 - 500 ms). Therefore, the anatomical model was made for one moment of the breathing phase. Because the anatomical model deforms due to breathing movements, we had to develop a model for calculating deformations based on the observation of the stomach surface. The process requires a lot of computing power, so we had to propose multithreaded data processing.

Results

The developed system can be helpful for surgeons in the process of removing liver tumors. It helps to estimate breathing movements and presents the position of tools relative to anatomical structures. The substantive manager of the project, with whom we continue to cooperate, received for the implementation of the project described above the prestigious "Teraz Polska" award and "Very Important Polish Innovator".