Objective To explore the application value of artificial intelligence (AI) pulmonary artery assisted diagnosis software for suspected pulmonary embolism patients. Methods The data of 199 patients who were clinically suspected of pulmonary embolism and underwent pulmonary artery CT angiography (CTA) from June 2016 to December 2021 were retrospectively analyzed. Images of pulmonary artery CTA diagnosed by radiologists with different experiences and judged by senior radiologists were compared with the analysis results of AI assisted diagnostic software for pulmonary artery CTA, to evaluate the diagnostic efficacy of this software and low, medium, and senior radiologists for pulmonary embolism. The agreement of pulmonary embolism based on pulmonary artery CTA between the AI software and radiologists with different experiences was evaluated using Kappa test. Results The agreement of the AI software and the evaluation of pulmonary embolism lesions by senior radiologists based on pulmonary artery CTA was high (Kappa=0.913, P<0.001), while the diagnostic results of pulmonary artery CTA AI software was good after judged by senior radiologists based on pulmonary artery CTA (Kappa=0.755, P<0.001). Conclusions The AI software based on pulmonary artery CTA diagnosis of pulmonary embolism has good consistency with diagnostic images of radilogists, and can save a lot of reconstruction and diagnostic time. It has the value of daily diagnosis work and worthy of clinical promotion.
ObjectiveTo explore the evaluation value of preoperative multislice spiral computed tomography angiography (MSCTA) for normative radical gastrectomy. MethodsThe anatomic distributions of celiac trunk and its three branches and their tributaries (common hepatic artery, right hepatic artery, left hepatic artery, splenic artery, and left gastric artery) of 86 patients with gastric cancer were comprehended by preoperative MSCTA, which were verified during the surgery. Simultaneously preoperative TNM staging was evaluated by MSCTA, which compared with postoperative pathological results. ResultsThe accuracy rate of preoperative MSCTA evaluating the distribution of celiac trunk and its three branches and their tributaries was 100%. Abnormal hepatic arteries were found in 22 cases by MSCTA, the mutation rate was 25.58%. Abnormal right hepatic arteries were found in 11 cases (12.79%), abnormal left hepatic arteries in 7 cases (8.14%), both abnormal right and left hepatic arteries in 1 case (1.16%), and abnormal common hepatic arteries in 3 cases (3.49%). Straight splenic arteries were found in 24 cases (27.91%), slightly curved splenic arteries in 44 cases (51.16%), and significantly curved splenic arteries in 18 cases (20.93%). Compared with postoperative pathological results, the accuracy rates of preoperative MSCTA evaluating gastric cancer T, N, and M staging were 75.58%（65/86）, 74.42%(64/86), and 91.86%(79/86), respectively. ConclusionsPreoperative MSCTA is an objective way to assess the distributions of celiac artery trunk and related tributaries of patients with gastric cancer. Also, it is an accurate method to evaluate the preoperative TNM stage of gastric cancer, which can help to make an individual operative plan and avoid the intraoperative injury of the artery.
Objective To investigate the influence of axis pedicle and intra-axial vertebral artery (IAVA) alignment on C2 pedicle screw placement by measuring the data of head and neck CT angiography. MethodsThe axis pedicle diameter (D), isthmus height (H), isthmus thickness (T), and IAVA alignment types were measured in 116 patients (232 sides) who underwent head and neck CT angiography examinations between January 2020 and June 2020. Defined the IAVA offset direction by referencing the vertical line through the center of C3 transverse foramen on the coronal scan, it was divided into lateral (L), neutral (N), and medial (M). Defined the IAVA high-riding degree by referencing the horizontal line through the outlet of the C2 transverse foramen, it was divided into below (B), within (W), and above (A). The rate of pedicle stenosis, high-riding vertebral artery, and different IAVA types were calculated, and their relationships were analysed. Simulative C2 pedicle screws were implanted by Mimics 19.0 software, and the interrelation among the rates of pedicle stenosis, high-riding vertebral artery, IAVA types, and vertebral artery injury were analyzed. ResultsThe rate of C2 pedicle stenosis was 33.6% (78/232), and the rate of high-riding vertebral artery was 35.3% (82/232). According to the offset direction and the degree of riding, IAVA was divided into 9 types, among which the N-W type (29.3%) was the most, followed by the L-W type (19.0%) and the L-B type (12.9%), accounting for 60.9%. The vertebral artery injury rate of simulative implanted C2 pedicle screws was 35.3% (82/232). The vertebral artery injury rate in patients with pedicle stenosis and high-riding vertebral artery was significantly higher than that who were not (P<0.001). The rate of pedicle stenosis, high-riding vertebral artery, and vertebral artery injury were significantly different among IAVA types (P<0.001), and M-A type was the most common. ConclusionVertebral artery injury is more common in pedicle stenosis and/or high-riding vertebral artery and/or IAVA M-A type. Preoperative head and neck CT angiography examination has clinical guiding significance.
Objective To explore the reliability and effectiveness of prediction of the pedicle length of the proximally-based anterolateral thigh (pALT) flap which was used to repair the defects following the resection of various malignant tumors using computed tomographic angiography (CTA). Methods The clinical data of 12 patients who met the selection criteria by using pALT flap to repair wounds left after malignant tumor resection between June 2015 and December 2020 were retrospectively analyzed. There were 5 males and 7 females; the age ranged from 16 to 80 years, with an average age of 54.4 years. After tumor resection, the soft tissue defect ranged from 15 cm×5 cm to 30 cm×12 cm; defect sites included 4 cases of lower abdomen, 3 cases of groin, 2 cases of thigh, and 3 cases of buttocks. Preoperative CTA was used to obtain the location information of the descending branch of the lateral femoral circumflex artery and its perforators by maximum density projection, and the length of the pedicle of pALT flap was estimated. Fasciocutaneous flap (5 cases) or myocutaneous flap (7 cases) were cut during operation to repair the defect, and the size of flap ranged from 20 cm×7 cm to 30 cm×12 cm. The donor site of thigh was directly sutured (11 cases) or repaired with skin graft (1 case). Bland-Altman analysis was used to detect the consistency between the pALT flap vascular pedicle length estimated by CTA and the pALT flap vascular pedicle length actually obtained during operation. ResultsOne case had distal blood supply disturbance of the flap and was repaired with skin graft after debridement; the remaining 11 flaps survived. All donor and recipient incisions healed by first intention. All 12 cases were followed up 1-12 months, with an average of 4.3 months. One patient died of pelvic tumor recurrence at 6 months after operation, and no tumor recurrence was found in the other patients. Preoperative CTA estimated that the length of pALT flap vascular pedicle was 9.3-24.7 cm, with an average of 14.7 cm; the actual length of pALT flap vascular pedicle was 9.5-25.0 cm, with an average of 14.8 cm. Bland-Altman analysis showed that there was no significant difference between the pALT flap vascular pedicle length estimated by CTA before operation and the pALT flap vascular pedicle length actually obtained during operation, and the average difference was 0.1 (95% consistency limit: –0.89, 0.74), indicating that they had good consistency. ConclusionCTA can be accurately used to localize the perforator and predict the possible pedicle length of the pALT flap. When performing a pALT flap surgery, preoperative CTA is helpful for surgeons to make a preliminary assessment of the difficult of the operation. The time for exploration of perforators and dissection of the vascular pedicle, and complications can be reduced, and the safety of the operation can be improved.
ObjectiveTo explore the feasibility of three-dimensional (3-D) visualization reconstruction of the medial sural artery perforator flap based on digital technology. MethodsA series of Dicom images were obtained from three healthy adult volunteers by dual source CT angiography. Then the Mimics software was used to construct the medial sural artery model and measure the indexes, including the starting position of medial sural artery, external diameters of vascular pedicle, the number of perforators, location perforated deep fascia, and the maximum pedicle length of perforators based on medial sural artery perforator flap. ResultsThe 3-D visualization reconstruction models were successfully finished with Mimics software, which can clearly display the distribution, travel, and perforating point. Thirteen perforators were found in 6 legs, which started at the popliteal artery with a mean external diameter of 2.3 mm (range, 1.9-2.7 mm). Each specimen had 1-3 perforators, which located at the site of 6.2-15.0 cm distal to popliteal crease and 2.5-4.2 cm from posterior midline. The maximum pedicle length of medial sural artery perforator flap was 10.2-13.8 cm (mean, 11.8 cm). ConclusionThe 3-D visualization reconstruction models based on digital technology can provide dynamic visualization of the anatomy of the medial sural artery for individualized design of the medial sural artery perforator flap.
ObjectiveTo investigate the effectiveness of digital technology in repairing wounds of the hand and foot with anterolateral thigh flap. MethodsBetween September 2013 and September 2014, 16 cases of wounds of the hand and foot were treated with the anterolateral thigh flap. There were 10 males and 6 females, with an average age of 31 years (range, 20-52 years). The causes included traffic accident injury in 8 cases, crushing injury by machine in 6 cases, burning injury in 1 case, and animal biting injury in 1 case. The locations of soft tissue defect were the dorsum of the foot in 5 cases, the ankle in 4 cases, the planta pedis in 1 case, and the hand and forearm in 6 cases. The time was 2 hours to 45 days from injury to hospitalization (mean, 14.3 days). All defects were associated with exposure of bone and tendon. The size of wound was from 9.0 cm×4.0 cm to 29.0 cm×8.5 cm. CT angiography (CTA) was performed before operation, and the appropriate perforator as well as the donor site was selected. Then the Mimics15.0 software was used to reconstruct the data of CTA so as to locate the main perforators, design the three-dimensional models of the anterolateral thigh flap, and simulate operation. The flap was obtained according to preoperative plan during operation. The size of flaps varied from 11 cm×5 cm to 31 cm×10 cm. The donor sites were sutured directly in 14 cases and were repaired by free skin graft in 2 cases. ResultsThe lateral femoral circumflex artery identified by Mimics15.0 software before operation, as well as the starting position of its descending branch, the blood vessel diameter at start site, vascular distribution, the maximum cutting length of the vascular pedicle were consistent with the actual observation during operation. All flaps were harvested and were used to repair defect smoothly. Vascular crisis occurred in 1 flap after operation, and the other flaps survived successfully. The wounds and the incisions obtained healing by first intention, and grafted skin survived completely. All cases were followed up 6-17 months (mean, 9 months). Fifteen flaps had good shape;but a second-stage operation was performed to make the flap thinner in 1 case. At last follow-up, the results were excellent in 3 cases, good in 2 cases, and fair in 1 case according to total active motion (TAM) in 6 cases of hand and forearm injury;the results were excellent in 5 cases, good in 3 cases, and fair in 2 cases according to American Orthopaedic Foot and Ankle Society (AOFAS) in 10 cases of foot injury. The total excellent and good rate was 81.25%. ConclusionThe preoperative individualization design of the flap can be realized through CTA digital technology and Mimics15.0 software;it can reduce the operation risk.
ObjectiveTo explore the feasibil ity to repair defect on the neck and chest with the rectus abdominis flap which pedicle is lengthened by measuring the width, thickness, and the intercostal space of the inferior costicartilage using CT angiography (CTA). MethodsThirty cases receiving CTA and three-dimensional reconstruction between July and December 2013 were included in the study. Of 30 cases, 17 were male and 13 were female, aged 44-70 years (mean, 56 years). The width and thickness of the 3rd to 7th costicartilages and the distance of the 3rd to 6th intercostal spaces were measured, and the lengthened pedicle was calculated after the 4th to 7th costicartilage was cut off. Between July 2012 and November 2013, the lengthened pedicle of the rectus abdominis flap was cl inically used to repair the defect on the neck and chest in 4 cases. ResultsThe pedicle of the rectus abdominis flap was about 6 cm in length. When the left 7th, 6th, 5th, and 4th costicartilages were cut off, the average pedicle was lengthened by 4.07, 7.99, 12.50, and 17.48 cm respectively; when the right 7th, 6th, 5th, and 4th costicartilages were cut off, the average pedicle was lengthened by 4.63, 10.82, 16.64, and 22.05 cm respectively. In 4 flaps which were cl inically used to repair defects, 3 flap completely survived, 1 flap had partial necrosis. Three patients were followed up 6 months, and the appearance and texture of the flap were satisfactory; 1 patient failed to be followed up. ConclusionResecting the inferior costicartilage can prolong the pedicle of the rectus abdominis flap, therefore it can be used to repair defect on the upper chest and the neck.
目的 探讨多层螺旋CT血管造影（CTA）在主动脉夹层中的诊断价值及临床应用。 方法 回顾性分析2010年2月－2011年4月35例行CTA检查的主动脉夹层患者，所有患者原始数据在图像后处理工作站采用多平面重建、容积再重建、最大密度投影等方法进行主动脉成像。由2名有经验的放射科副主任医师进行诊断。 结果 35例均可明确显示主动脉夹层的真假腔、内膜片及破裂口部位。Ⅰ型12例，Ⅱ型3例，Ⅲ型20例；累及左锁骨下动脉5例，左颈总动脉2例，无名动脉2例，腹腔干3例，肠系膜上动脉4例，左肾动脉3例，右肾动脉2例，右髂总动脉受累6例，左髂总动脉受累8例，其中双侧髂总动脉均受累4例；合并动脉瘤3例；壁内血肿4例；所有患者均显示了单一或多发破口。 结论 CTA及图像后处理技术能快速、准确地诊断主动脉夹层，为临床治疗方案选择提供重要的影像学依据。
Objective To evaluate the effectiveness of the innervated medial plantar flap for reconstructing soft tissue defects, particularly in the weight-bearing zone, after resection of foot tumors. MethodsA retrospective analysis was conducted on 12 patients with malignant skin and soft tissue tumors of the foot treated between October 2023 and December 2024. The cohort included 8 males and 4 females, aged 42-67 years (mean, 57.5 years). Tumor types comprised malignant melanoma (5 cases), squamous cell carcinoma (4 cases), arsenical keratosis (2 cases), and tumor-induced osteomalacia (1 case). Soft tissue defects located in the heel weight-bearing region in 10 cases and non-weight-bearing ankle region in 2 cases, with defect sizes ranging from 4.0 cm×3.0 cm to 6.0 cm×4.0 cm. Preoperative photon-counting CT angiography (PC-CTA) was performed to assess the medial plantar artery and its perforators. All patients underwent radical tumor resection with confirmed negative margins. The resulting defects were reconstructed using a innervated medial plantar flap incorporating sensory branches of the medial plantar nerve. The flap donor site was covered with a split-thickness skin graft harvested from the ipsilateral inguinal region. Results The operation was successfully completed in all 12 patients. All flaps survived completely without vascular compromise, partial necrosis, or total loss. Incisions healed primarily without dehiscence or infection. Minor skin graft necrosis occurred at the donor site in 3 patients, which healed within 2-3 weeks with routine dressing changes. No donor site complication (e.g., tendon or nerve injury) occurred. Patients were followed up 2-16 months (mean, 10.3 months). At last follow-up, there was no tumor recurrence. Flaps exhibited good color and texture match with surrounding tissue, restored sensation, and all feet achieved normal weight-bearing activity. Conclusion The innervated medial plantar flap, precisely designed based on PC-CTA localization, provides reliable blood supply and effective sensory restoration. It is an ideal method for reconstructing soft tissue defects after foot tumor resection, especially in the heel weight-bearing region.
Objective To investigate the clinical value of computed tomographic angiography (CTA) and three-dimensional reconstruction technique in repairing scalp avulsion wound with large skull exposure by the free latissimus dorsi flap transplantation. Methods Between October 2007 and June 2012, 9 female patients with serious scalp avulsion and large skull exposure were treated, aged 23-54 years (mean, 38 years). The injury causes included machine twist injury in 6 cases, traffic accident injury in 2 cases, and falling from height injury in 1 case. Before admission, 3 patients had scalp necrosis after scalp in situ replantation, and 6 patients underwent debridement and dressing. The time from injury to admission was 8 hours to 7 days (mean, 1 day). The avulsed scalp area ranged from 75% to 90% of the scalp area (mean, 81%); the exposed skull area ranged from 55% to 70% of the scalp area (mean, 63%). Two patients had unilateral auricle avulse. CTA was used to observe the superficial temporal artery and vein, facial artery, external jugular vein, dorsal thoracic artery and vein, and measure the blood vessel diameter before operation. According to the CTA results, the latissimus dorsal skin flaps were desinged to repair wounds in 7 cases, the latissimus dorsal muscle flaps combined with skin graft were used to repair wounds in 2 cases. According to preoperative design, operation was successfully completed in 7 cases; great saphenous vein was used as vascular graft in 2 cases having poor images of superficial temporal vessels. The size of latissimus dorsal skin flaps ranged from 20 cm × 14 cm to 25 cm × 20 cm; the donor site was repaired with skin graft. The size of latissimus dorsal muscle flaps were 23 cm × 16 cm and 16 cm × 10 cm; the donor site was directly sutured. Results The blood vessel diameter measured during operation was close to the value measured before operation. The operation time was 6-8 hours (mean, 6.5 hours). The latissimus dorsal muscle (skin) flap and skin graft survived, with primary healing of wound or incision at donor site. The patients were followed up 3 months-2 years (mean, 6 months). The flap had soft texture and skin had no ulceration. Conclusion The free latissimus dorsi flaps can repair scalp avulsion with large skull exposure. Preoperative CTA can get the vessel anatomical structure and diameter at donor and recipient sites, which will guide the operation program design and implementation so as to shorten the operation time and improve the accuracy rate of vascular anastomosis.