Overview
Introduction
k-Plan is a software package for planning transcranial ultrasound procedures, including transcranial ultrasound stimulation (TUS). It uses a streamlined workflow to select an ultrasound device, position the device relative to a subject using a template or medical image, and specify the sonication parameters. Calculations of the ultrasound field and temperature inside the skull and brain are then automatically executed in the cloud. k-Plan avoids the need for detailed knowledge of numerical modelling or access to high-performance computing resources in order to run planning simulations.
The best way to get started with k-Plan is to work through the quick start guide. This covers many of the key features of k-Plan. Then, work through each of the topics in the user guide to gain a deeper understanding of how the software operates.
k-Plan modules
k-Plan consists of three modules:
- Planning module (TPM): The TPM is a GUI-based application installed on the end-user's computer. The TPM is used to define and manage subject data (such as name and age) and plan data (such as the transducer position and sonication parameters). All the k-Plan user data is stored locally as part of the TPM installation. Subject and plan data are stored locally in an encrypted database, while plan images are copied to the local k-Plan data folder.
- Dispatch server module (DSM): The DSM is a dispatch service (called k-Dispatch) running in the cloud. The TPM installations communicate with k-Dispatch to automatically schedule planning simulations to run on remote high-performance computing servers. k-Dispatch is also used to verify license and user credentials, to manage cloud computing allocations, and to provide access to the TPM software installer for end-users. Users can login to the k-Dispatch web interface at https://dispatch.kplan.io.
- Simulation execution module (SEM): The SEM executes the planning simulations on remote (cloud-based) computing hardware using the k-Wave simulation software. The SEM takes the plan details, and converts this to a series of k-Wave simulations (see simulation pipeline). The results are then processed, and stored in a results file which is downloaded by the TPM via the DSM when the simulation is complete.
Data flow
To create a plan, the end user uses the k-Plan TPM to add medical image data and select associated parameters (for example, the position of the ultrasound device relative to the head, and the sonication parameters).
When the plan is complete, the medical image data along with the therapy settings are saved locally into a planning file in a proprietary file format. The planning file is pseudonymised and does not contain any identifiable information except for the raw medical image data. The code which links the planning file with the subject data is only stored locally within the encrypted database that forms part of the end-user’s installation of the TPM.
When the end-user selects to evaluate the plan, the planning file is transmitted from the TPM to the DSM using secure communication protocols. The DSM then schedules the plan for execution by the SEM. When the simulation is complete, the results are saved into a results file in a proprietary file format. This file contains information about the predicted ultrasound and thermal exposures for the plan.
The end-user can query the status of the simulation via the TPM. When the simulation is complete, the end-user can select to download the results file from the DSM to their TPM installation using secure communication protocols. The results are then displayed within the TPM for review by the end-user. Optionally, the control parameters can also be exported to an external ultrasound device.
The k-Plan DSM and SEM modules are not a data storage system. The planning file is sent from the end-user’s TPM and temporarily held on the DSM server while being processed. After the simulation, the results file is temporarily held on the DSM server before being sent back to the end-user’s TPM for long-term retention. Both files are then securely deleted from the DSM.
The k-Plan DSM and SEM use servers that are located with the United Kingdom and the European Union. Access to the DSM and SEM servers is protected and restricted to authorised administrative users.
Data hierarchy
The k-Plan TPM connects subjects and plans using a structured database as described below.
- Subjects: At the highest level is the subject. This generally corresponds to an individual participant in a study, but can also be used to group other plans, for example, a set of template simulations. Each subject can be identifiable or anonymised. In both cases, subject and study IDs must be specified. These are retained if an identifiable subject within k-Plan is later anonymised by selecting the
Anonymisedcheckbox. - Plans: Each subject can have one or more plans. Each plan is connected to a specific transducer, a set of planning images, and a set of simulation parameters. A plan generally corresponds to a single study.
- Sonications: Each plan can have one or more sonications. In k-Plan, each sonication is defined by a transducer position, the transducer parameters (such as frequency and focal length), and the pulse timing parameters. Different sonications can be used within a plan, for example, to target different brain regions, or for investigating the effects of different sonication parameters for a single target.
Major dialogs
The k-Plan TPM contains three main dialogs that the mirror the data hierarchy outlined above.
- Subject browser dialog: Used to add, modify, and subjects.
- Plan browser dialog: Used to add, modify, and select plans.
- Plan developer dialog: Used to setup the plan, including adding sonications.
Details of how to create new subjects and plans are given in the quick start guide.
