Objective To explore six-degree-of-freedom (6-DF) registration methods between planning and cone beam computed tomography (CBCT) during image-guided radiation therapy (IGRT) in esophageal cancer.Methods Thirty pairs of CBCT images acquired before radiation and the corresponding planning computed tomography (CT) images of esophageal cancer were selected for further investigation.Registration markers for 6-DF image registration were determined and contoured in those images.The results of registration as well as time cost were compared among different registration methods of bone match, gray value match, manual match, and bone plus manual match.Results Contouring bone and spinal canal posterior to the target volume of esophageal carcinoma as registration marker could make 6-DF registration quick and precise.Compared with manual match, set-up errors of v rotation in bone plus manual match (-0.55° vs.-0.88°, t=2.55, P=0.020), of x-axis and v rotation in bone match (0.12 mm vs.-2.33 mm, t=5.75, P=0.000; -0.35° vs.-0.88°, t=3.00, P=0.007), and of x-axis and w rotation in gray value match (7.20 mm vs.-2.33 mm, t=3.10, P=0.006; -0.10° vs.-0.59°, t=2.81, P =0.011) were significantly different.Compared with manual match, the coincidence rate of bone plus manual match was the highest (85.55%), followed by bone match and gray value match (74.45% and 74.45%).The time cost of each registration method from longest to shortest was:6.00 -10.00 minutes for manual match, 1.00 - 5.00 minutes for bone plus manual match, 0.75 - 1.50 minutes for gray value match, and 0.50 - 0.83 minutes for bone match.Conclusions Registration marker is useful for image registration of CBCT and planning CT in patients with esophageal cancer.Bone plus manual match may be the best registration method considering both registration time and accuracy.

Objective To investigate the age-related pattern of normal skull bone marrow with 3. 0 T MR T1WI. Methods Cranial MR T1WI images which were defined to be normal were retrospectively reviewed in 360 cases. Patients with known diffuse bone marrow disease, focal lesions, history of radiation treatment or steroid therapy were excluded, while patients whose cranial MRI and follow-up visits were all normal were included in this study. All the subjects were divided into 7 groups according to the age: < 1, 1--2,3--5, 6--14, 15--29, 30--49, >50 years group. Mid- and para- sagittal T1WI images were used to be analyzed and the type of cranial bone marrow was classified according to the thickness of diploe and the pattern of the signal characteristics. Statistical analysis was conducted to reveal the relationship between the age and the type. Results The normal skull bone marrow could be divided into four types as follows: (1) Type- Ⅰ : 115 cases, 47 of which appeared type- Ⅰ a and the mean thickness was ( 1.24±0. 31 ) mm; 68 of which appeared type- Ⅰ b and the mean thickness was ( 1.76 ± 0. 37 ) mm. Type- Ⅱ : 57 cases and the mean thickness was (2.78 ±0.69) mm. Type-Ⅲ: 148 cases, 18 of which appeared type-Ⅲ a and the mean thickness was (2. 33±0. 65) mm; 88 of which appeared type-Ⅲ b and the mean thickness was (4. 01 ± 0. 86) mm; 42 of which appeared type-Ⅲ c and the mean thickness was (4. 31±0. 73) mm. Type-Ⅳ: 40 cases, 25 of which appeared type-Ⅳ a and the mean thickness was (5. 17 ± 1.02) mm; 15 of which appeared type-Ⅳ b and the mean thickness was (5.85±1.45) mm. (2) <1 year group: 40 cases, 20 of which appeared type- Ⅰ a, 20 type- Ⅰ b and the mean thickness of this group was ( 1.47 ± 0. 42 ) mm. 1-- 2 years group: 40 cases, 16 of which appeared type- Ⅰ a, 18 type- Ⅰ b, 6 type- Ⅱ and the mean thickness of this group was ( 1.68±0. 52) mm. 3--5 years group: 40 cases, 8 of which appeared type- Ⅰ a, 18 type-Ⅰb, 14 type-Ⅱ and the mean thickness of this group was (1.84±0.73) mm. 6--14 years group: 60 cases, 3 of which appeared type- Ⅰ a, 12 type- Ⅰ b, 27 type-Ⅱ , 12 type-Ⅲb, 6 type-Ⅲc, and the mean thickness of this group was (2.92±1. 00) mm. 15--29 years group: 60 cases, 7 of which appeared type-Ⅱ , 5 type-ma, 27 type-Ⅲb, 15 type-Ⅲc, 6 type-Ⅳa and the mean thickness of this group was (3.95 ± 0.97) mm. 30---49 years group: 60 cases, 3 of which appeared type-Ⅱ , 7 type-Ⅲa, 29 type-Ⅲb, 15 type-Ⅲc, 3 type-Ⅳa and 3 type-Ⅳb and the mean thickness of this group was (4. 30 ± 1.35) mm. ≥50 years group: 60 cases, 6 of which appeared type-Ⅲ a, 20 type-Ⅲb, 6 type-Ⅲc, 16 type-Ⅳa and 12 type-Ⅳb and the mean thickness of this group was (4.51 ± 1.40) mm. (3) There is a linear relationship between ages and types that is revealed by chi-square test (x2 = 266. 36, P < 0. 01 ). Conclusion There is characteristic in the distribution of normal skull bone marrow with age growing. And skull bone marrow transforms gradually from type- Ⅰ to Ⅳ with aging.

Objective To investigate the morbidity, diagnostic profile and perinatal outcome of pregestational diabetes mellitus (PGDM) in 15 hospitals in Guangdong province. Methods A total of 41338 women delivered in the 15 hospitals during the 6 months,195 women with PGDM(PGDM group) and 195 women with normal glucose test result(control group)were recruited from these tertiary hospitals in Guangdong province from January 2016 to June 2016. The morbidity and diagnostic profile of PGDM were analyzed. The complications during pregnancy and perinatal outcomes were compared between the two groups. In the PGDM group, pregnancy outcomes were analyzed in women who used insulin treatment (n=91) and women who did not (n=104). Results (1)The incidence of PGDM was 0.472%(195/41338). Diabetes mellitus were diagnosed in 59 women (30.3%, 59/195) before pregnancy, and 136 women (69.7%,136/195) were diagnosed as PGDM after conceptions. Forty-six women (33.8%) were diagnosed by fasting glucose and glycohemoglobin (HbA1c) screening. (2) The maternal age, pre-pregnancy body mass index (BMI), prenatal BMI, percentage of family history of diabetes, incidence of macrosomia, concentration of low density lipoprotein were significantly higher in PGDM group than those in control group (all P<0.05). Women in PGDM group had significantly higher HbA1c concentration((6.3±1.3)% vs (5.2±0.4)%), fasting glucose [(6.3±2.3) vs (4.8±1.1) mmol/L], oral glucose tolerance test(OGTT)-1 h glucose((12.6±2.9) vs (7.1± 1.3) mmol/L)and OGTT-2 h glucose [(12.0±3.0) vs (6.4±1.0) mmol/L] than those in control group (P<0.01). (3)The morbidity of preterm births was significantly higher (11.3% vs 1.0%, P<0.01), and the gestational age at delivery in PGDM group was significantly smaller [(37.6±2.3) vs (39.2±1.2) weeks, P<0.01]. Cesarean delivery rate in the PGDM group (70.8% vs 29.7%) was significantly higher than the control group (P<0.01). There was significantly difference between PGDM group and control in the neonatal male/female ratio (98/97 vs 111/84, P=0.033). The neonatal birth weight in PGDM group was significantly higher((3159±700) vs (3451±423) g, P<0.01). And the incidence of neonatal hypoglycemia in the PGDM group was higher than the control group (7.7% vs 2.6%, P=0.036).(4)In the PGDM group, women who were treated with insulin had a smaller gestational age at delivery [(36.9±2.9) vs (37.9±2.5) weeks, P<0.01], and the neonates had a higher neonatal ICU(NICU)admission rate (24.2% vs 9.6% , P<0.01). Conclusions The morbidity of PGDM in the 15 hospitals in Guangdong province is 0.472%. The majority of PGDM was diagnosed during pregnancy; HbA1c and fasting glucose are reliable parameters for PGDM screening. Women with PGDM have obvious family history of diabetes and repeated pregnancy may accelerate the process of diabetes mellitus. Women with PGDM have higher risk for preterm delivery and neonatal hypoglycemia. Unsatisfied glucose control followed by insulin treatment may increase the need for NICU admission.

Insulinoma is a rare neuroendocrine tumor originating from pancreatic beta cells, which clinically manifests Whipple′ s triad and recurrent hypoglycemia. Insulinoma in a patient with type 2 diabetes mellitus is even a more rarely encountered case. In clinical practice, hypoglycemia in patients with diabetes is often considered to be associated with oral hypoglycemic agents, but insulinoma, as a possible etiology, is usually neglected. This article reported a case of a patient with type 2 diabetes mellitus, before the diagnosis of which she had experienced recurrent hypoglycemia for about 14 years. Hypoglycemia symptoms with hyperinsulinemia kept existing even after the withdrawal of antidiabetic agents. The pancreatic imaging and postoperative histopathology supported the diagnosis of insulinoma. After surgery, plasma glucose level of the patient increased and diabetes treatments continued. The co-existence of insulinoma and type 2 diabetes mellitus in this case may be attributed to the insulin resistance induced by chronic hyperinsulinemia due to insulinoma.