Clinical Validation of a Deep-Learning Segmentation Software in Head and Neck: An Early Analysis in a Developing Radiation Oncology Center.

BACKGROUND: Organs at risk (OARs) delineation is a crucial step of radiotherapy (RT) treatment planning workflow. Time-consuming and inter-observer variability are main issues in manual OAR delineation, mainly in the head and neck (H & N) district. Deep-learning based auto-segmentation is a promising strategy to improve OARs contouring in radiotherapy departments. A comparison of deep-learning-generated auto-contours (AC) with manual contours (MC) was performed by three expert radiation oncologists from a single center. METHODS: Planning computed tomography (CT) scans of patients undergoing RT treatments for H&N cancers were considered. CT scans were processed by Limbus Contour auto-segmentation software, a commercial deep-learning auto-segmentation based software to generate AC. H&N protocol was used to perform AC, with the structure set consisting of bilateral brachial plexus, brain, brainstem, bilateral cochlea, pharyngeal constrictors, eye globes, bilateral lens, mandible, optic chiasm, bilateral optic nerves, oral cavity, bilateral parotids, spinal cord, bilateral submandibular glands, lips and thyroid. Manual revision of OARs was performed according to international consensus guidelines. The AC and MC were compared using the Dice similarity coefficient (DSC) and 95% Hausdorff distance transform (DT). RESULTS: A total of 274 contours obtained by processing CT scans were included in the analysis. The highest values of DSC were obtained for the brain (DSC 1.00), left and right eye globes and the mandible (DSC 0.98). The structures with greater MC editing were optic chiasm, optic nerves and cochleae. CONCLUSIONS: In this preliminary analysis, deep-learning auto-segmentation seems to provide acceptable H&N OAR delineations. For less accurate organs, AC could be considered a starting point for review and manual adjustment. Our results suggest that AC could become a useful time-saving tool to optimize workload and resources in RT departments.

Dosimetric Impact of Inter-Fraction Variability in the Treatment of Breast Cancer: Towards New Criteria to Evaluate the Appropriateness of Online Adaptive Radiotherapy.

Purpose: As a discipline in its infancy, online adaptive RT (ART) needs new ontologies and ad hoc criteria to evaluate the appropriateness of its use in clinical practice. In this experience, we propose a predictive model able to quantify the dosimetric impact due to daily inter-fraction variability in a standard RT breast treatment, to identify in advance the treatment fractions where patients might benefit from an online ART approach. Methods: The study was focused on right breast cancer patients treated using standard adjuvant RT on an artificial intelligence (AI)-based linear accelerator. Patients were treated with daily CBCT images and without online adaptation, prescribing 40.05 Gy in 15 fractions, with four IMRT tangential beams. ESTRO guidelines were followed for the delineation on planning CT (pCT) of organs at risk and targets. For each patient, all the CBCT images were rigidly aligned to pCT: CTV and PTV were manually re-contoured and the original treatment plan was recalculated. Various radiological parameters were measured on CBCT images, to quantify inter-fraction variability present in each RT fraction after the couch shifts compensation. The variation of these parameters was correlated with the variation of V95% of PTV (ΔV95%) using the Wilcoxon Mann-Whitney test. Fractions where ΔV95% > 2% were considered as adverse events. A logistic regression model was calculated considering the most significant parameter, and its performance was quantified with a receiver operating characteristic (ROC) curve. Results: A total of 75 fractions on 5 patients were analyzed. The body variation between daily CBCT and pCT along the beam axis with the highest MU was identified as the best predictor (p = 0.002). The predictive model showed an area under ROC curve of 0.86 (95% CI, 0.82-0.99) with a sensitivity of 85.7% and a specificity of 83.8% at the best threshold, which was equal to 3 mm. Conclusion: A novel strategy to identify treatment fractions that may benefit online ART was proposed. After image alignment, the measure of body difference between daily CBCT and pCT can be considered as an indirect estimator of V95% PTV variation: a difference larger than 3 mm will result in a V95% decrease larger than 2%. A larger number of observations is needed to confirm the results of this hypothesis-generating study.

Predicting respiratory distress syndrome in neonates from mothers with intrahepatic cholestasis of pregnancy.

OBJECTIVE: Intrahepatic cholestasis of pregnancy (ICP) has been associated with prematurity and fetal mortality. Recently, ICP has also been recognised as a risk factor for neonatal respiratory distress syndrome (RDS) in term or near-term neonates. Since fetal mortality is more frequent in pregnancies with an early ICP onset, we speculated that the time of exposure (ET) to maternal bile acids at the delivery (BAdeliv) could be involved in neonatal lung damage too. Study aim was to develop a scoring system to predict the RDS occurrence. DESIGN: We conducted a retrospective analysis of 77 pregnancies complicated by ICP (years 2000-2004) looking for factors associated to the neonatal RDS. We developed a risk score as follows: RDS risk score=BAdeliv x ET/gestational age and we prospectively applied it to 30 neonates from ICP pregnancies (years 2005-2006). RESULTS: ROC analysis indicated 9 as the score with the highest sensitivity (83.3%) and specificity (87.5%). Considering a RDS incidence of about 25% in babies coming from ICP pregnancies, the post-test probability showed a risk increased to 66.7% with a score>9 and reduced to 4.8% with a score

Using BiliCheck for preterm neonates in a sub-intensive unit: diagnostic usefulness and suitability.

BACKGROUND: BiliCheck (BC), a new transcutaneous bilirubinometer is thought to be lacking in the disadvantages of old devices and could be potentially useful for diagnosing jaundice in preterm babies. Although its accuracy is well known in healthy term babies, there is a lack of knowledge about its usefulness in preterm infants. AIMS: To investigate BC usefulness in preterm babies and its suitability in a sub-intensive neonatal unit. STUDY DESIGN: In 340 preterm infants between 30 and 36 weeks of gestational age, transcutaneous and serum bilirubin measurement were performed. Hematocrit, pH, postnatal age, gestational age, and sex were also studied to clarify their influence on BC accuracy. For a subset of 100 neonates transcutaneous measurement, blood collection and serum analysis were timed and costs were considered. RESULTS: Correlation coefficient is 0.795 (p<0.001) and this is not affected by factors previously supposed to be important. Overall sensitivity was 100% and specificity were comprised between 40% and 72%. BC has a tendency to overestimate serum bilirubin, at high values. Considering the whole time for serum bilirubin measurement, transcutaneous bilirubinometry is a faster (p<0.0001), but more expensive technique with a cost of about 5 euro/measurement. Nevertheless, using BC as a screening-device we could safely avoid 58-79% of blood samples, since its positive predictive values is about 21-42%. This would allow to a cost reduction of 1555-2120 euro/year. CONCLUSIONS: BC has a good reliability in preterm infants although not as good as in healthy term babies. BC is a time-sparing tool and can improve the management of neonatal jaundice in preterm infants; however, its tendency to overestimate suggests its use only for screening purposes.

Delivery room strategies and outcomes in preterm infants with gestational age 24-28 weeks.

OBJECTIVE: To investigate the effect of different delivery room strategies on survival, short term morbidity, and outcomes in extremely premature infants. METHODS: This retrospective cohort study included all preterm infants with a gestational age between 24 and 28 weeks who were born in 1992-1997 (period A; n = 161) and in 1998-2003 (period B; n = 163). In period A, elective intubation was performed. In period B, if spontaneous breathing was present, nasal continuous positive airway pressure (nCPAP) was applied. RESULTS: Survival rate and the number of never-intubated infants significantly increased in period B. No differences were found concerning short-term morbidity. Among major outcomes, the need for retinopathy of prematurity (ROP) surgery and the length of stay were significantly lower in period B. Subgroup analysis showed no significant differences from period A to period B in infants with gestational age 24-26 weeks. In the 27-28 weeks subgroup, the never-intubated infants rate increased from 2.8% to 21.3% and survival rate increased from 63% to 79%. A reduced need for ROP surgery and a shorter hospital stay were also observed. CONCLUSIONS: Changes in delivery room strategy tending to reduce mechanical ventilation in extremely premature infants are likely to benefit essentially infants of 27-28 weeks of gestation. Extension of such benefits to premature infants at the limit of viability requires further research.

Intrahepatic cholestasis of pregnancy and neonatal respiratory distress syndrome.

OBJECTIVES: We sought to verify the association between maternal intrahepatic cholestasis of pregnancy (ICP) and neonatal respiratory distress syndrome (RDS) and to determine how bile acids levels alter the risk of developing neonatal RDS. METHODS: We extracted data from our divisional database about all of the newborns born during the years 2000-2004. We compared 77 neonates born from pregnancies complicated by ICP with 427 neonates in the same range of gestational age born from noncomplicated pregnancies. We studied maternal bile acids levels immediately before delivery in mothers with ICP and measured bile acid levels during the first 24 hours of life in their newborns. RESULTS: The incidence of RDS in newborns from cholestatic pregnancies was twice that the reference population (28.6% vs 14%). The multivariate analysis showed that the risk of RDS in these newborns was approximately 2.5 times higher than in control infants. Within the ICP group, maternal and neonatal bile acid levels of infants affected by RDS were not significantly higher than those of healthy infants. The multivariate analysis showed that a low gestational age was the most important risk factor, but the probability of respiratory distress syndrome also increased by 2 per thousand for every additional micromole of the interaction term "neonatal by maternal bile acids level." CONCLUSIONS: Maternal ICP is significantly associated with the occurrence of RDS in the newborn. We hypothesize that bile acids can produce surfactant depletion in the alveoli reverting the reaction of phospholipase A2. This hypothesis could potentially be confirmed by bronchoalveolar lavage study.