Patient-Specific Electric Field Simulation In Spinal Metastasis Electrochemotherapy

How a protocol simulates patient-specific electric fields during spinal electrochemotherapy and compare them with post-treatment mononucleosis. Existing workflows rarely validate simulated electric fields against three-dimensional treatment response in spinal metastatic disease. After importing the data into the 3D Slicer and setting CTI as the reference volume, open the segment editor module and create a segment named cortical bone.

Use the threshold tool in CTI to isolate high density cortical bone. Then manually refine the segment using the scissors or paint tools as needed. Create a segment named Fat.

Use the Threshold tool to select low density fat in the perirenal or subcutaneous space. Next, construct an intervertebral disc segment and use the 3D Paint tool on axial, sagittal, and coronal views. Select outside all visible segments to restrict painting to empty voxels and avoid overriding existing segments.

Then set up a cancellous bone segment for vertebral cancellous bone. Use the 3D Paint tool to fill the internal trabecular compartments, and select outside all visible segments as described previously. Create a segment labeled Spinal Cord and use the 3D Paint tool to manually contour the cord on axial slices from above to below the treated levels.

Next, define a segment named cerebrospinal fluid. Use the 3D Paint tool to fill the cerebrospinal fluid space surrounding the spinal cord within the thecal sac, and also select outside all visible segments to avoid overriding existing segments. Create additional segments for cement, coils, or lungs if present.

Use the threshold tool to identify hyperdense cement or coils and hypodense lung parenchyma. Perform manual correction as needed. Then save the updated Slicer scene.

Import the pre-procedural MRI volume, MRIp, into the Slicer scene. Use the Transforms module to perform rigid registration of MRIp onto CTI using vertebral osseous landmarks at the treated level to guide registration. Apply the Transform to MRIp and visually assess co-registration in axial and sagittal planes.

Accept the registration only if the visual alignment and target registration error are less than 2.5 millimeters. Then create a new segment named Tumor. Use the Paint tool on MRIp to manually delineate epidural and vertebral tumor involvement at baseline.

Apply optional smoothing to the tumor segment to obtain a coherent three-dimensional volume without gaps. And save the updated Slicer scene. Import the intraprocedural computed tomography volume with needle CTA into the 3D Slicer scene.

Utilize the transforms module for rigid registration of CTA onto CTI. Employ the same vertebral bony landmarks as in the previous step to ensure proper alignment. Launch the AI4DEEP module and create one virtual needle for each electrode, starting from the active electrode tip.

Verify in axial, coronal, and sagittal views that each electrode number matches accurately with a single physical active tip. Save the updated scene containing anatomical structures, tumor segmentation, and electrogeometry. In the AI4DEEP interface, assign each anatomical segment to a corresponding tissue class with predefined electrical conductivity.

Set the default background tissue to muscle conductivity so that all non-segmented voxels are treated as muscle. Verify in the AI4DEEP summary panel that each segment is aligned with the expected tissue type and conductivity. Then save the configuration.

In the AI4DEEP interface, specify the active tip length for each electrode according to the clinical procedure and set the electrodiameter to 1.8 millimeters. Enter the clinically applied electric field amplitude in volts per centimeter. Then define the electrode pairs that were activated during the treatment.

Let AI4DEEP compute the applied voltage for each electrode pair based on the inter-electrode distance while ensuring the requested voltage to distance ratio in volts per centimeter is respected. Set the number of pulses to eight and the pulse duration to 100 microseconds according to the clinical protocol. Then save the AI4DEEP configuration.

In the AI4DEEP interface, start the electrostatic field computation using the configured anatomy, conductivities, and electrode parameters. Once the software generates the geometrical configuration and solves the linear electrostatic potential, save the simulation output and the updated Slicer scene. In the AI4DEEP interface, select the option to generate electric field isosurfaces from the simulated field.

Select a range of isodose thresholds and generate the corresponding three-dimensional isodose maps. Display individual isodose maps in a three-dimensional view to observe the spatial field distribution. The field typically ranges from pale yellow at low amplitudes to red at high amplitudes.

For each isodose, record the percentage of tumor coverage values automatically computed based on the overlap between the isodose volume and the tumor segmentation. Save all isodose segments or models and the tumor coverage values for subsequent analysis. Perform follow-up MRI at six weeks after electrochemotherapy and every two months thereafter.

Select the MRI scan showing the largest necrotic volume or best radiological response for segmentation to account for a heterogeneous and delayed treatment response. Import the post-treatment contrast enhanced T1 fat-suppressed axial sequence centered on the treated levels MRIs into 3D Slicer. Perform rigid registration of MRIs onto CTI using the same predefined vertebral landmarks as used previously.

Apply the Transform. Calculate the target registration error using the same reference landmarks, and confirm that it is less than or equal to 2.5 millimeters to document registration accuracy. Create a new segment named Tumor necrosis.

Use the Paint tool to manually delineate the non-enhancing necrotic portion of the tumor on the post-electrochemotherapy MRI slice by slice. Apply optional smoothing to the tumor necrosis segment to obtain a coherent contiguous volume. Then perform a quantitative comparison between the simulated field and necrosis.

Mean Dice similarity coefficients showed a bell-shaped relationship across the cohort. The values peaked between 160 and 200 volts per centimeter and tumor coverage gradually declined as thresholds increased. Marked heterogeneity was observed between cases with best Dice values ranging from 0.0156 to 0.7684 and corresponding best Iso values ranging from 100 to 500 volts per centimeter.

A representative overlay showed spatial correspondence between post-treatment tumor necrosis and the 200 volt per centimeter isodose volume. A 59-year-old patient with L3 epidural metastasis from cholangiocarcinoma showed less than 5%tumor necrosis and no improvement after the initial electrochemotherapy. Retrospective simulation with AI4DEEP demonstrated inadequate tumor coverage at the 200 volts per centimeter threshold.

A repeat treatment with an adjusted electrode configuration achieved over 90%tumor coverage, resulting in a complete radiologic and clinical response. A 67-year-old patient with clear cell renal cell carcinoma and L5-S1 epidural disease developed right-sided radiculopathy after electrochemotherapy. AI4DEEP simulation showed 300 volts per centimeter field extension into the right S1 foramen, consistent with postoperative MRI-confirmed right S1 nerve root injury.

The main challenge is identifying the relevant isodose threshold for each treatment, organ, and clinical setting. Additional analysis can test automatic segmentation, registration uncertainty, and dose-response relationships between electric field thresholds. Further studies should validate these early findings in larger spinal metastasis cohorts and other electro chemotherapy indications.