Comparison of breath-hold and free-breathing positions of an external fiducial by analysis of respiratory traces

Respiratory signal prediction based on adaptive boosting and multi-layer perceptron neural network

To improve the prediction accuracy of respiratory signals using adaptive boosting and multi-layer perceptron neural network (ADMLP-NN) for gated treatment of moving target in radiation therapy. The respiratory signals acquired using a real-time position management (RPM) device from 138 previous 4DCT scans were retrospectively used in this study. The ADMLP-NN was composed of several artificial neural networks (ANNs) which were used as weaker predictors to compose a stronger predictor. The respiratory signal was initially smoothed using a Savitzky-Golay finite impulse response smoothing filter (S-G filter). Then, several similar multi-layer perceptron neural networks (MLP-NNs) were configured to estimate future respiratory signal position from its previous positions. Finally, an adaptive boosting (Adaboost) decision algorithm was used to set weights for each MLP-NN based on the sample prediction error of each MLP-NN. Two prediction methods, MLP-NN and ADMLP-NN (MLP-NN plus adaptive boosting), were evaluated by calculating correlation coefficient and root-mean-square-error between true and predicted signals. For predicting 500 ms ahead of prediction, average correlation coefficients were improved from 0.83 (MLP-NN method) to 0.89 (ADMLP-NN method). The average of root-mean-square-error (relative unit) for 500 ms ahead of prediction using ADMLP-NN were reduced by 27.9%, compared to those using MLP-NN. The preliminary results demonstrate that the ADMLP-NN respiratory prediction method is more accurate than the MLP-NN method and can improve the respiration prediction accuracy.

Lung volume reproducibility under ABC control and self-sustained breath-holding

An Active Breathing Coordinator (ABC) can be employed to induce breath-holds during CT imaging and radiotherapy of lung, breast and liver cancer, and recently during lung cancer MRI. The apparatus measures and controls respiratory volume, hence subject lung volume reproducibility is its principal measure of effectiveness. To assess ABC control quality, the intra-session reproducibility of ABC-induced lung volumes was evaluated and compared with that reached by applying the clinical standard of operator-guided self-sustained breath-holds on healthy volunteers during MRI. Inter-session reproducibility was investigated by repeating ABC-controlled breath-holds on a second visit. Additionally, lung volume agreement with ABC devices used with different imaging modalities in the same institution (MR, CT), or for a breast trial treatment, was assessed. Lung volumes were derived from three-dimensional (3D) T1-weighted MRI datasets by three observers employing semiautomatic lung delineation on a radiotherapy treatment planning system. Inter-observer variability was less than 6% of the delineated lung volumes. Lung volume agreement between the different conditions over all subjects was investigated using descriptive statistics. The ABC equipment dedicated for MR application exhibited good intra-session and inter-session lung volume reproducibility (1.8% and 3% lung volume variability on average, respectively). MR-assessed lung volumes were similar using different ABC equipment dedicated to MR, CT, or breast radiotherapy. Overall, lung volumes controlled by the same or different ABC devices agreed better than with self-controlled breath-holds, as suggested by the average ABC variation of 1.8% of the measured lung volumes (99 mL), compared to the 4.1% (226 mL) variability observed on average with self-sustained breath-holding.

Fiducial marker placement via conventional or electromagnetic navigation bronchoscopy (ENB): an interdisciplinary approach to the curative management of lung cancer

BACKGROUND AND AIMS

Conventional and electromagnetic navigation bronchoscopy (ENB) is generally used as a diagnostic tool in suspicious pulmonary nodules. The use of this technique for the placement of fiducial markers in patients with inoperable but early-stage lung cancer could present an innovative approach enabling risk-reduced therapy.

METHODS

We present seven clinical cases where conventional bronchoscopy and ENB were used as part of an experimental interdisciplinary approach to clinical management and therapy planning. In each case, we analyzed the clinical indication, endoscopic procedures and post-interventional outcome.

RESULTS

In six patients (three females, three males) with peripheral non-small cell lung cancer (NSCLC), stage cT1cN0cM0, surgery and conventional stereotactic radiation therapy was not possible because of end-stage chronic obstructive pulmonary disease. ENB was used for fiducial marker placement prior to cyberknife radiotherapy. No procedure-related complications were observed. Complete remission could be achieved in four cases, partial remission in two cases and no relevant complications induced by radiotherapy were observed. In one male patient, an endoluminal relapse in the right lower lobe was diagnosed following a right upper lobe resection for a NSCLC. The tumor could not be clearly identified by computerized tomography, so that the bronchoscopic placement of a fiducial marker in the tumor was performed in order to allow stereotactic radiochemotherapy, by which complete remission could be achieved.

CONCLUSION

Fiducial marker placement may be an interesting bronchoscopic technique in the interdisciplinary therapeutic approach to inoperable early-stage lung cancer. In the described cases, therapy planning was successful and no procedure-related complications were observed.

Respiratory movement of upper abdominal organs and its effect on radiotherapy planning in pancreatic cancer

AIMS

Radiotherapy for pancreatic cancer is complicated by the frequent overlapping of the planning target volume (PTV) and the organ at risk (OAR), limiting the dose that can be safely delivered to the tumour. Individualising the margins applied to the clinical target volume (CTV) may reduce OAR irradiation without increasing the risk of geographical miss. We quantified the movement of the pancreas with respiration and evaluated whether individualised margins based on this motion reduced the dose to OARs.

MATERIALS AND METHODS

Planning computed tomography scans were acquired in quiet breathing, held expiration and held inspiration. Organ motion was evaluated from displacement of a reproducible point within the pancreas in all directions. Two sets of plans (standard plan: P(stan); individualised plan incorporating movement data: P(ind)) were generated for each patient. The PTV and doses to OARs were evaluated for both sets of plans.

RESULTS

The mean (standard deviation) movement of the pancreas in the superior-inferior, lateral and anterior-posterior directions were 15.3 mm (4.3), 5.2 mm (3.5) and 9.7 mm (6.1), respectively. The use of individualised margins reduced the mean PTV volume by 33.5% (9.8) (P=0.0051). The proportional reductions in the percentage of kidney receiving >10 Gy, small bowel >45 Gy and liver >30 Gy were 63.7% (P=0.0051), 29.3% (P=0.0125) and 29.2% (P=0.0107), respectively. For the same level of OAR constraints, individualised margins allowed dose escalation in six of the 10 patients to a mean dose of 63.2 Gy.

CONCLUSIONS

The present study shows a simple way of incorporating organ motion into the planning process and can be adopted by any centre without major strain on healthcare resources. The use of individualised margins reduced PTV volume and the dose to OARs. This may offer an opportunity for dose escalation to try and further improve local control.