ABSTRACT
Objective:Due to the low contrast between tumors and surrounding tissues in CBCT images, this study was designed to propose an automatic segmentation method for central lung cancer in CBCT images.Methods:There are 221 patients with central lung cancer were recruited. Among them, 176 patients underwent CT localization and 45 patients underwent enhanced CT localization. The enhanced CT images were set as the lung window and mediastinal window, and elastic registration was performed with the first CBCT validation images to obtain the paired data set. The CBCT images could be transformed into" enhanced CT" under the lung window and mediastinal window by loading the paired data sets into cycleGAN network for style transformation. Finally, the transformed images were loaded into the UNET-attention network for deep learning of GTV. The results of segmentation were evaluated by Dice similarity coefficient (DSC), Hausdorff distance (HD) and the area under the receiver operating characteristic curve (AUC).Results:The contrast between tumors and surrounding tissues was significantly improved after style transformation. The DSC value of cycleGAN+ UNET-attention network was 0.78±0.05, HD value was 9.22±3.42 and AUC value was 0.864, respectively.Conclusion:The cycleGAN+ UNET-attention network can effectively segment central lung cancer in CBCT images.
ABSTRACT
<p><b>OBJECTIVE</b>To explore the clinical and genetic characteristics of a case with Pallister-Killian syndrome (PKS).</p><p><b>METHODS</b>Chromosomal karyotype of umbilical cord blood sample derived from a 36-year-old pregnant woman was analyzed by G-banding analysis. After birth, the child was further analyzed with single nucleotide polymorphism microarray (SNP array) and fluorescence in situ hybridization (FISH) using 12pter/12qter probes.</p><p><b>RESULTS</b>G-banding analysis showed that the fetus has a karyotype of 46,XY [77]/47,XY,+mar [23]. After birth, Affymetrix CytoScan 750K array analysis showed a segmental tetrasomy of arr [hg19] 12p13.33p11.1(173 786 - 34 835 641)×4 and a 34.6 Mb repeat at 12p13.33p11.1 with in the neonate. FISH analysis confirmed that 39% of cells harbored the 12p tetrasomy.</p><p><b>CONCLUSION</b>Combined clinical examination, G-banded chromosomal karyotyping, FISH and microarray analysis can delineate the origin and fragments of small supernumerary marker chromosomes and diagnose PKS with precision.</p>
Subject(s)
Adult , Female , Humans , Pregnancy , Chromosome Banding , Chromosome Disorders , Diagnosis , Chromosomes, Human, Pair 12 , In Situ Hybridization, Fluorescence , Methods , Karyotyping , Oligonucleotide Array Sequence Analysis , Polymorphism, Single Nucleotide , Prenatal Diagnosis , MethodsABSTRACT
Objective To investigate the setup error for nasopharyngeal carcinoma (NPC) patients with poor compliance using kV cone-beam computed tomography,and to calculate the expansion margin from the clinical target volume (CTV) to planning target volume (PTV).Methods In 45 NPC patients from 2013 to 2015,the setup error,95％ confidence interval (CI)-1 for random error,and PTV-1 value were calculated.Moreover,in 16 NPC patients with poor compliance based on five verifications (random error not within 95％ CI-1),the setup error,95％ CI-2 for random error,and PTV-2 value were calculated.For the 16 special patients,PTV-1 and PTV-1 combined with PTV-2 were used to develop the plan-1 and plan-2,respectively.The dosimetric difference between plan-1 and plan-2 was evaluated.Results Both PTV-1 and PTV-2 had the largest expansion margin in the y direction.The CTV of plan-1 could not meet the requirement of the prescription dose after the setup error was introduced.Compared with plan-1,the V95％ and D95 values for the CTV of plan-2 were increased by 6.26％ and 4.43％,respectively.The D01 value was significantly larger in plan-2 than in plan-1 (P=0.005),which,however,met the clinical requirement.Conclusions In patients with poor compliance,the dose to target volume can be effectively elevated and the normal tissue can be spared from damage when PTV-1 combined with PTV-2 is selected as expansion margin.
ABSTRACT
Objective To analyze the relation between glycosylated hemoglobin (HbA1c) and coronary heart disease (CHD) among diabetic and non-diabetic individuals.Methods A total of 1190 in patients from January 2009 to September 2009 were selected and divided into two groups:225 patients with diabete mellitus and 965 patients without.We collected the age,sex,triglyceride ( TG),total cholesterol ( CHO),high-density lipoprotein ( HDL-C),low-density lipoprotein ( LDL-C),fasting blood-glucose (FBG),glycosylated hemoglobin (HbA1c),high-sensitivity CRP (hs-CRP) and analyzed the relation between these indexes and CHD in diabetic and non-diabetic individuals respectively,then,analyzed the difference of HbA1c between patients with and without CHD after correcting the differentiating factor with a multivariable-adjusted model.Meanwhile,according the HbA1c level,we divided all participants into four groups:4.0％ -5.4％( Ⅰ),5.5％ -5.9％ ( Ⅱ),6.0％ -6.4％ ( Ⅲ),≥6.5％ ( Ⅳ),and observed the distribution of HbA1c and coronary heart disease in diabetic and non-diabetic individuals respectively.Results ( 1) In non-diabetic individuals,statistically significant difference of male( 80.5％ υs 62.7％),age [ ( 59 ±11) years υs (55 ± 11) years],FBG [(5.62 ± 1.48) mmol/L υs (5.30 ±0.84) mmol/L],HbA1c [(5.98±0.92)％ υs (5.65 ±0.53)％],CHO [(4.48 ±1.01) mmol/L υs (4.77 ±1.04) mmol/L],LDL-C [(2.59±0.87) mmol/L υs (2.79 ±0.86) mmol/L],HDL-C [(1.08 ±0.26) mmol/L υs(1.21 ±0.32) mmol/L] was observed between patients with and without CHD (P ＜0.01),however,in diabetic individuals,only LDL-C [ ( 2.56 ± 0.81) mmol/L υs ( 3.07 ± 0.90) mmol/L],CHO [ (4.44 ±0.95) mmol/L υs ( 5.08 ± 1.16) mmol/L] were observed ( P ＜ 0.0 1).(2) In all participants,the higher of HbA1c,the higher of CHD percent,compared to patients without CHD,the percent of high HbA1c in CHD patients was significantly higher.After adjusting for age,sex,HDL-C,HbA1c was an independent risk marker for CHD in non-diabetic individuals ( OR:1.935,95％ CI:1.356 - 2.762,P ＜ 0.05),however,links between FBG in the non-diabetic range and CHD appeared weaker after controlling indexes listed above( OR:1.507,95％CI:1.082 -2.098,P＜0.05).Conclusion HbA1c is an independent risk factor for CHD in nondiabetic individuals and prior to FBG.