Automatic Lung Segmentation and Quantification of Aeration in Computed Tomography of the Chest Using 3D Transfer Learning

Effects of Positive End-expiratory Pressure on Pulmonary Perfusion Distribution and Intrapulmonary Shunt during One-lung Ventilation in Pigs: A Randomized Crossover Study

BACKGROUND

During one-lung ventilation (OLV), positive end-expiratory pressure (PEEP) can improve lung aeration but might overdistend lung units and increase intrapulmonary shunt. The authors hypothesized that higher PEEP shifts pulmonary perfusion from the ventilated to the nonventilated lung, resulting in a U-shaped relationship with intrapulmonary shunt during OLV.

METHODS

In nine anesthetized female pigs, a thoracotomy was performed and intravenous lipopolysaccharide infused to mimic the inflammatory response of thoracic surgery. Animals underwent OLV in supine position with PEEP of 0 cm H2O, 5 cm H2O, titrated to best respiratory system compliance, and 15 cm H2O (PEEP0, PEEP5, PEEPtitr, and PEEP15, respectively, 45 min each, Latin square sequence). Respiratory, hemodynamic, and gas exchange variables were measured. The distributions of perfusion and ventilation were determined by IV fluorescent microspheres and computed tomography, respectively.

RESULTS

Compared to two-lung ventilation, the driving pressure increased with OLV, irrespective of the PEEP level. During OLV, cardiac output was lower at PEEP15 (5.5 ± 1.5 l/min) than PEEP0 (7.6 ± 3 l/min) and PEEP5 (7.4 ± 2.9 l/min; P = 0.004), while the intrapulmonary shunt was highest at PEEP0 (PEEP0: 48.1% ± 14.4%; PEEP5: 42.4% ± 14.8%; PEEPtitr: 37.8% ± 11.0%; PEEP15: 39.0% ± 10.7%; P = 0.027). The relative perfusion of the ventilated lung did not differ among PEEP levels (PEEP0: 65.0% ± 10.6%; PEEP5: 68.7% ± 8.7%; PEEPtitr: 68.2% ± 10.5%; PEEP15: 58.4% ± 12.8%; P = 0.096), but the centers of relative perfusion and ventilation in the ventilated lung shifted from ventral to dorsal and from cranial to caudal zones with increasing PEEP.

CONCLUSIONS

In this experimental model of thoracic surgery, higher PEEP during OLV did not shift the perfusion from the ventilated to the nonventilated lung, thus not increasing intrapulmonary shunt.

EDITOR’S PERSPECTIVE

Precision of CT-derived alveolar recruitment assessed by human observers and a machine learning algorithm in moderate and severe ARDS

BACKGROUND

Assessing measurement error in alveolar recruitment on computed tomography (CT) is of paramount importance to select a reliable threshold identifying patients with high potential for alveolar recruitment and to rationalize positive end-expiratory pressure (PEEP) setting in acute respiratory distress syndrome (ARDS). The aim of this study was to assess both intra- and inter-observer smallest real difference (SRD) exceeding measurement error of recruitment using both human and machine learning-made lung segmentation (i.e., delineation) on CT. This single-center observational study was performed on adult ARDS patients. CT were acquired at end-expiration and end-inspiration at the PEEP level selected by clinicians, and at end-expiration at PEEP 5 and 15 cmH₂O. Two human observers and a machine learning algorithm performed lung segmentation. Recruitment was computed as the weight change of the non-aerated compartment on CT between PEEP 5 and 15 cmH₂O.

RESULTS

Thirteen patients were included, of whom 11 (85%) presented a severe ARDS. Intra- and inter-observer measurements of recruitment were virtually unbiased, with 95% confidence intervals (CI_95%) encompassing zero. The intra-observer SRD of recruitment amounted to 3.5 [CI_95% 2.4-5.2]% of lung weight. The human-human inter-observer SRD of recruitment was slightly higher amounting to 5.7 [CI_95% 4.0-8.0]% of lung weight, as was the human-machine SRD (5.9 [CI_95% 4.3-7.8]% of lung weight). Regarding other CT measurements, both intra-observer and inter-observer SRD were close to zero for the CT-measurements focusing on aerated lung (end-expiratory lung volume, hyperinflation), and higher for the CT-measurements relying on accurate segmentation of the non-aerated lung (lung weight, tidal recruitment…). The average symmetric surface distance between lung segmentation masks was significatively lower in intra-observer comparisons (0.8 mm [interquartile range (IQR) 0.6-0.9]) as compared to human-human (1.0 mm [IQR 0.8-1.3] and human-machine inter-observer comparisons (1.1 mm [IQR 0.9-1.3]).

CONCLUSIONS

The SRD exceeding intra-observer experimental error in the measurement of alveolar recruitment may be conservatively set to 5% (i.e., the upper value of the CI_95%). Human-machine and human-human inter-observer measurement errors with CT are of similar magnitude, suggesting that machine learning segmentation algorithms are credible alternative to humans for quantifying alveolar recruitment on CT.

Segmental Lung Recruitment in Patients with Bilateral COVID-19 Pneumonia Complicated by Acute Respiratory Distress Syndrome: A Case Report

Bilateral COVID-19 pneumonia is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and usually leads to life-threatening acute respiratory distress syndrome (ARDS). Treatment of patients with ARDS is difficult and usually involves protective mechanical ventilation and various types of recruitment maneuvers. A segmental lung recruitment maneuver by independent lung ventilation has been described as a successful recruitment maneuver in patients with lobar pneumonia, and may, therefore, be useful for the treatment of patients with bilateral COVID-19 pneumonia complicated by ARDS in the critical phase of the disease when all other therapeutic options have been exhausted. The aim of this case series was to present a case report of four mechanically ventilated patients with severe bilateral COVID-19 pneumonia complicated by ARDS using the segmental lung recruitment maneuver. The effect of the segmental lung recruitment maneuver was assessed by the increase in PaO₂/FiO₂ ratio and the lung ultrasound (LUS) scoring system (0 points-presence of sliding lungs with A-lines or one or two isolated B-lines; 1 point-moderate loss of lung ventilation with three to five B lines; 2 points-severe loss of lung ventilation with more than five B lines (B pattern); and 3 points-lung consolidation) determined 12, 24, and 48 h after segmental lung recruitment. In three of four patients with bilateral COVID-19 pneumonia complicated by ARDS, an increase in the PaO₂/FiO₂ ratio and an improvement in the LUS scoring system were observed 48 h after segmental lung recruitment. In conclusion, the segmental lung recruitment maneuver in patients with bilateral COVID-19 complicated by ARDS is an effective method of lung recruitment and may be a useful treatment method.

Advanced Analysis of 3D Kinect Data: Supervised Classification of Facial Nerve Function via Parallel Convolutional Neural Networks

In this paper, we designed a methodology to classify facial nerve function after head and neck surgery. It is important to be able to observe the rehabilitation process objectively after a specific brain surgery, when patients are often affected by face palsy. The dataset that is used for classification problems in this study only contains 236 measurements of 127 patients of complex observations using the most commonly used House–Brackmann (HB) scale, which is based on the subjective opinion of the physician. Although there are several traditional evaluation methods for measuring facial paralysis, they still suffer from ignoring facial movement information. This plays an important role in the analysis of facial paralysis and limits the selection of useful facial features for the evaluation of facial paralysis. In this paper, we present a triple-path convolutional neural network (TPCNN) to evaluate the problem of mimetic muscle rehabilitation, which is observed by a Kinect stereovision camera. A system consisting of three modules for facial landmark measure computation and facial paralysis classification based on a parallel convolutional neural network structure is used to quantitatively assess the classification of facial nerve paralysis by considering facial features based on the region and the temporal variation of facial landmark sequences. The proposed deep network analyzes both the global and local facial movement features of a patient’s face. These extracted high-level representations are then fused for the final evaluation of facial paralysis. The experimental results have verified the better performance of TPCNN compared to state-of-the-art deep learning networks.