Laparoscopic Projection Mapping of the Liver Portal Segment, Based on Augmented Reality Combined With Artificial Intelligence, for Laparoscopic Anatomical Liver Resection.

Hepatocellular carcinoma causes intrahepatic metastasis via the trans-portal vein. Thus, appropriate mapping of portal segments is necessary for laparoscopic anatomical liver resection. However, because of the difficulty in identifying tactile sensations and the limited surgical view of laparoscopy, augmented reality (AR) has recently been utilized in laparoscopic liver surgery to identify the tumor, vessels, and portal segments. Moreover, artificial intelligence (AI) has been employed to identify landmarks in two-dimensional (2D) images because of concerns regarding the accuracy of superimposing a three-dimensional (3D) model onto a 2D laparoscopic image. In this study, we report an AR-based projection mapping method of portal segments superimposing preoperative 3D models assisted by AI in laparoscopic surgery. The liver silhouette in laparoscopic images should be detected to superimpose 3D models. Labeled liver silhouettes were obtained from 380 images in surgical videos as learning images to implement AI-based silhouette detection. To implement this technique, we used Detectron2, a PyTorch-based object detection library by Facebook AI Research (Now, Meta AI, Menlo Park, California, United States). In the videos, the liver edges were displayed as green outlines according to AI. Additionally, 3D liver models with segmental mapping were generated using the open-source software 3D Slicer from computed tomography images. For AR display, we utilized the model target function of Vuforia SDK (PTC, Inc., Boston, Massachusetts, United States), an industrial AR library with silhouette-based AR display. Lastly, we merged the AI output video with a 3D model in Unity (Unity Software Inc., San Francisco, California, United States) to establish the projection mapping of the portal segment on 2D surgical images. The accuracy was assessed by measuring the maximum error between the liver edges of laparoscopic images and 3D liver silhouettes in five surgical videos. The maximum error between liver edges and 3D model silhouettes ranged from 4 mm to 22 mm in the AI-based approach and 12 mm to 55 mm in the non-AI-based approach. Meanwhile, the mean error was 14.5 and 31.2 mm in the AI-based and non-AI-based approaches, respectively. Despite camera movement, 3D AR displays were maintained. Thus, our AI-assisted projection mapping of the portal segment could offer a new approach for laparoscopic anatomical liver resection.

Unobtrusive Refractive Power Monitoring: Using EOG to Detect Blurred Vision.

The rise in population and aging has led to a significant increase in the number of individuals affected by common causes of vision loss. Early diagnosis and treatment are crucial to avoid the consequences of visual impairment. However, in early stages, many visual problems are making it difficult to detect. Visual adaptation can compensate for several visual deficits with adaptive eye movements. These adaptive eye movements may serve as indicators of vision loss. In this work, we investigate the association between eye movement and blurred vision. By using Electrooculography (EOG) to record eye movements, we propose a new tracking model to identify the deterioration of refractive power. We verify the technical feasibility of this method by designing a blurred vision simulation experiment. Six sets of prescription lenses and a pair of flat lenses were used to create different levels of blurring effects. We analyzed binocular movements through EOG signals and performed a seven-class classification using the ResNet18 architecture. The results revealed an average classification accuracy of 94.7% in the subject-dependent model. However, the subject-independent model presented poor performance, with the highest accuracy reaching only 34.5%. Therefore, the potential of an EOG-based visual quality monitoring system is proven. Furthermore, our experimental design provides a novel approach to assessing blurred vision.

Identification of Language-Induced Mental Load from Eye Behaviors in Virtual Reality.

Experiences of virtual reality (VR) can easily break if the method of evaluating subjective user states is intrusive. Behavioral measures are increasingly used to avoid this problem. One such measure is eye tracking, which recently became more standard in VR and is often used for content-dependent analyses. This research is an endeavor to utilize content-independent eye metrics, such as pupil size and blinks, for identifying mental load in VR users. We generated mental load independently from visuals through auditory stimuli. We also defined and measured a new eye metric, focus offset, which seeks to measure the phenomenon of "staring into the distance" without focusing on a specific surface. In the experiment, VR-experienced participants listened to two native and two foreign language stimuli inside a virtual phone booth. The results show that with increasing mental load, relative pupil size on average increased 0.512 SDs (0.118 mm), with 57% reduced variance. To a lesser extent, mental load led to fewer fixations, less voluntary gazing at distracting content, and a larger focus offset as if looking through surfaces (about 0.343 SDs, 5.10 cm). These results are in agreement with previous studies. Overall, we encourage further research on content-independent eye metrics, and we hope that hardware and algorithms will be developed in the future to further increase tracking stability.

Effect of duration of adjuvant chemotherapy with S-1 (6 versus 12 months) for resected pancreatic cancer: the multicenter clinical randomized phase II postoperative adjuvant chemotherapy S-1 (PACS-1) trial.

BACKGROUND: Six-month adjuvant chemotherapy with S-1 is standard care for resected pancreatic cancer in Japan; however, the optimal duration has not been established. We aimed to evaluate the impact of duration of adjuvant chemotherapy with S-1. METHODS: We performed a multicenter, randomized, open-label, phase II study. Patients with histologically proven invasive pancreatic ductal carcinoma, pathological stage I-III, and no local residual or microscopic residual tumor were eligible. Patients were randomized 1:1 to receive 6- or 12-month adjuvant chemotherapy with S-1. The primary endpoint was 2-year overall survival (OS). Secondary endpoints were disease-free survival (DFS) and feasibility. RESULTS: A total of 170 patients were randomized (85 per group); the full analysis set was 82 in both groups. Completion rates were 64.7% (6-month group) and 44.0% (12-month group). Two-year OS was 71.5% (6-month group) and 65.4% (12-month group) (hazard ratio (HR): 1.143; 80% confidence interval CI 0.841-1.553; P = 0.5758). Two-year DFS was 46.4% (6-month group) and 44.9% (12-month group) (HR: 1.069; 95% CI 0.727-1.572; P = 0.6448). In patients who completed the regimen, 2-year DFS was 56.5% (6-month group) and 75.0% (12-month group) (HR: 0.586; 95% CI 0.310-1.105; P = 0.0944). Frequent (≥ 5%) grade ≥ 3 adverse events comprised anorexia (10.5% in the 6-month group) and diarrhea (5.3% vs. 5.1%; 6- vs. 12-month group, respectively). CONCLUSIONS: In patients with resected pancreatic cancer, 12-month adjuvant chemotherapy with S-1 was not superior to 6-month therapy regarding OS and DFS.

Effects of AR-Based Home Appliance Agents on User's Perception and Maintenance Behavior.

Maintenance of home appliances can be tedious. Maintenance work can be physically demanding and it is not always easy to know the cause of a malfunctioning appliance. Many users need to motivate themselves to perform maintenance work and consider it ideal for home appliances to be maintenance-free. On the other hand, pets and other living creatures can be taken care of with joy and without much pain, even if they are difficult to take care of. To alleviate the hassle associated with the maintenance of home appliances, we propose an augmented reality (AR) system to superimpose an agent over the home appliance of concern who changes their behavior according to the internal state of the appliance. Taking a refrigerator as an example, we verify whether such AR agent visualization motivates users to perform maintenance work and reduces the associated discomfort. We designed a cartoon-like agent and implemented a prototype system using a HoloLens 2, which can switch between several animations depending on the internal state of the refrigerator. Using the prototype system, a Wizard of Oz user study comparing three conditions was conducted. We compared the proposed method (Animacy condition), an additional behavior method (Intelligence condition), and a text-based method as a baseline for presenting the refrigerator state. In the Intelligence condition, the agent looked at the participants from time to time as if it was aware of them and exhibited help-seeking behavior only when it was considered that they could take a short break. The results show that both the Animacy and Intelligence conditions induced animacy perception and a sense of intimacy. It was also evident that the agent visualization made the participants feel more pleasant. On the other hand, the sense of discomfort was not reduced by the agent visualization and the Intelligence condition did not improve the perceived intelligence or the sense of coercion further compared to the Animacy condition.

Femoropopliteal Bypasses with Varicose Great Saphenous Vein.

The great saphenous vein is the conduit of choice for femoropopliteal or infrapopliteal bypass, but it is traditionally recommended that varicose vein grafts (VVGs) should not be used for bypass conduits owing to the risk of immediate rupture or long-term aneurysmal change. Herein, we report two cases of femoropopliteal bypass with VVGs. They achieved primary patency without aneurysmal formation after 32 and 17 months. Therefore, VVGs without morphologically conspicuous abnormalities are worth considering for usage as a vein graft.

Impact of a 7.5-Fr Pancreatic Stent for Preventing Pancreatic Fistula after Pancreaticoduodenectomy.

INTRODUCTION: Pancreatic duct stents are widely used to reduce the incidence of postoperative pancreatic fistula (POPF) after pancreaticoduodenectomy (PD); however, small stents may cause adverse effects, such as occlusion. Recently, we have tried placing a 7.5-Fr pancreatic duct stent to achieve more effective exocrine output from the pancreas; however, the association between pancreatic duct stent size and POPF remains unknown. METHODS: Sixty-five patients with soft pancreatic texture who underwent PD were retrospectively analyzed. After dividing the pancreas, a pancreatic duct stent (stent size 4.0 in 29 patients, 5.0 in 18, and 7.5 Fr in 18) was placed in the main pancreatic duct. RESULTS: Twenty-five of 65 patients with soft pancreatic texture (38.5%) developed POPF. POPF became less frequent as the pancreatic duct stent size increased (p = 0.003). The factors associated with POPF development were a 7.5-Fr pancreatic duct stent (p = 0.005), 5.0-Fr pancreatic duct stent (p = 0.031), and male sex (p = 0.008). Pancreatic duct stent size and pancreatic duct diameter did not differ between the POPF and non-POPF groups. DISCUSSION/CONCLUSIONS: In patients with a soft pancreas, the placement of a 7.5-Fr pancreatic duct stent may reduce the incidence of POPF.

Efficacy of neoadjuvant chemotherapy in patients with high-risk resectable colorectal liver metastases.

BACKGROUND: The role of preoperative neoadjuvant chemotherapy (NAC) in patients with resectable colorectal liver metastases (CRLM) remains undetermined. This study aimed to assess the efficacy of NAC in patients with resectable CRLM, especially in high-risk subgroups for recurrence, with special reference to synchronicity and the CRLM grade in the Japanese classification system. METHODS: A retrospective analysis of a multi-institutional cohort who was diagnosed with resectable CRLM was performed. CRLM was classified into three grades (A, B, and C) according to the combination of H stage (H1: ≤ 4 lesions and ≤ 5 cm, H2: ≥ 5 lesions or > 5 cm, H3: ≥ 5 lesions and > 5 cm), nodal status of the primary tumor (pN0/1: ≤ 3 metastases, pN2: ≥ 4 metastases), and the presence of resectable extrahepatic metastases. RESULTS: Among 222 patients with resectable CRLM, 97 (43.7%) had synchronous CRLM. The surgical failure-free survival (SF-FS) of patients with synchronous CRLM (without NAC) was significantly worse than that of patients with metachronous CRLM (P = 0.0264). The SF-FS of patients with Grade B/C was also significantly worse than that of Grade A (P = 0.0058). Among the 53 patients with synchronous and Grade B/C CRLM, 31 were assigned to NAC, and all of them underwent liver surgery. In this high-risk subgroup, the SF-FS and OS in the NAC group were significantly better than those in the upfront surgery group (P < 0.0001 and P = 0.0004, respectively). CONCLUSIONS: Patients with synchronous and Grade B/C CRLM could be good candidates for indication of NAC.

Prognostic significance of preoperative PNI and CA19-9 for pancreatic ductal adenocarcinoma: A multi-institutional retrospective study.

BACKGROUND: The aim of this study was to investigate the clinical value of nutritional and immunological prognostic scores as predictors of outcomes and to identify the most promising scoring system for patients with pancreatic ductal adenocarcinoma (PDAC) in a multi-institutional study. METHODS: Data were retrospectively collected for 589 patients who underwent surgical resection for PDAC. Prognostic analyses were performed for overall (OS) and recurrence-free survival (RFS) using tumor and patient-related factors, namely neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, Prognostic Nutritional Index (PNI), Glasgow Prognostic Score (GPS), modified GPS, C-reactive protein-to-albumin ratio, Controlling Nutritional Status score, and the Geriatric Nutritional Risk Index. RESULTS: Compared with PDAC patients with high PNI values (≥46), low PNI (<46) patients showed significantly worse overall survival (OS) (multivariate hazard ratio (HR), 1.432; 95% CI, 1.069-1.918; p = 0.0161) and RFS (multivariate HR, 1.339; 95% CI, 1.032-1.736; p = 0.0277). High carbohydrate antigen 19-9 (CA19-9) values (≥450) were significantly correlated with shorter OS (multivariate HR, 1.520; 95% CI, 1.261-2.080; p = 0.0002) and RFS (multivariate HR, 1.533; 95% CI, 1.199-1.961; p = 0.0007). Stratification according to PNI and CA19-9 was also significantly associated with OS and RFS (log rank, P < 0.0001). CONCLUSIONS: Our large cohort study showed that PNI and CA19-9 were associated with poor clinical outcomes in PDAC patients following surgical resection. Additionally, combining PNI with CA19-9 enabled further classification of patients according to their clinical outcomes.

Resection and reconstruction of pancreatic artery aneurysms caused by the compression of the celiac trunk by the median arcuate ligament: a report of two cases.

BACKGROUND: Some patients with the compression of the celiac trunk by the median arcuate ligament (MAL) suffer pancreatic artery aneurysms (PAAs) due to excessive blood flow from the superior mesenteric artery. These aneurysms are in peril because they are prone to rupture irrespective of size. Here, we present two cases of resection and reconstruction of PAAs caused by the compression of the celiac trunk by the MAL. CASE PRESENTATION: Patient 1 was a 44-year-old man who was first diagnosed to have a visceral artery aneurysm with a diameter of 4 cm accidentally found by ultrasound examination at a regular medical check-up. Contrast-enhanced CT revealed the compression of the celiac trunk by the MAL and a PAA originating from the first jejunal artery. First, laparoscopic excision of the MAL followed by a stent placement into the celiac trunk was performed. Although the stent was patent, the PAA still grew. The patient underwent resection and reconstruction of the PAA. Reconstruction of the pancreatic arterial arcade was needed because clamping of the inferior pancreaticoduodenal artery (IPDA) resulted in disappearance of the hepatic arterial blood flow. The follow-up CT 2 years and 9 months after the operation revealed no recurrence of aneurysms and the patent anastomosis. Patient 2 was a 68-year-old man who presented with an epigastric pain. Contrast-enhanced CT revealed the compression of the celiac trunk by the MAL and a PAA approximately 6 cm in diameter originating from the IPDA. The PAA was surrounded by a relatively low-intensity area, suggesting impending rupture of the PAA. The patient underwent resection and reconstruction of the PAA under an emergency situation. Reconstruction of the pancreatic arterial arcade was needed because clamping of the inflow IPDA resulted in disappearance of the hepatic blood flow. The follow-up CT 1 year and 8 months after the operation revealed no recurrence of aneurysms and the patent anastomosis. CONCLUSIONS: Although long-term follow-up is needed, resection and reconstruction is one of the therapeutic choices for PAAs caused by the compression of the celiac trunk by the MAL in order to prevent catastrophic aneurysm rupture.

Impact of Osteopenia on Oncologic Outcomes After Curative Resection for Pancreatic Cancer.

BACKGROUND/AIM: We evaluated the relationship between low bone mineral density (BMD), also called osteopenia, and prognosis in patients who underwent resection for pancreatic cancer (PC). PATIENTS AND METHODS: We enrolled 91 consecutive patients who underwent curative resections for PC between May 2009 and January 2019. Their BMDs were measured at the Th11 vertebra using computed tomography. Patients were then divided by age-adjusted standard BMD values into the osteopenia group (n=34) and the non-osteopenia group (n=57). Their overall survival (OS) and recurrence-free survival (RFS) were compared (log-rank test). RESULTS: The two groups did not differ in age, BMI, tumor marker, operation time, blood loss, postoperative complications or stage. The osteopenia group had significantly worse 3-year rates for OS (46% vs. 30%, p=0.04) and RFS (41% vs. 26%, p=0.01). In multivariate analysis, osteopenia was an independent prognostic factor for RFS (HR=2.16, p=0.01). CONCLUSION: Osteopenia is an adverse prognostic factor for patients with resected PC.

Validating the efficacy of interval appendectomy for acute appendicitis: representative three cases with different etiologies.

BACKGROUND: Appendectomy for acute appendicitis (AA) is considered one of the most common emergency surgeries. However, emergency appendectomy accompanied with complex lesions such as extensive abscess formation is not recommended in most cases. Therefore, non-operative management followed by interval appendectomy (IA) for AA has been tried. Herein, we present three AA cases with specific etiology that underwent interval appendectomy. CASE PRESENTATION: Case 1: A 68-year-old man was diagnosed AA with intestinal malrotation and intra-abdominal abscesses. He initially treated with conservative therapy and underwent laparoscopic IA after detailed preoperative examination. Case 2: A 22-year-old man had been under treatment for pancolitis-type ulcerative colitis (UC), also bothered by right-lower abdominal pain several times a year. The appendix always appeared swollen on every CT taken during symptoms. He underwent laparoscopic IA; pathological finding revealed typical UC histological features in the resected appendix. After the surgery, he never suffered from terrible right lower abdominal pain. Case 3: A 69-year-old woman complaining a right lower abdominal pain had undergone CT examination, which revealed AA with appendiceal mass, irregular wall thickness of the cecum, and mediastinal and para-aortic lymph node swelling. The operation was carried out after conservative therapy. The pathological diagnosis revealed BRAF mutated colorectal carcinoma. She had received systematic chemotherapy after the surgery, and all metastatic lesions have completely disappeared. CONCLUSION: Interval appendectomy provided us with much clearer anatomical information and precise therapeutic strategies, avoiding technical and general operative complications, and also induced fast recovery and short length of hospital stay. Interval appendectomy is a reasonable procedure and could be recommended in case of AA with some different etiology.

Impact of middle hepatic artery reconstruction after living donor liver transplantation using the left lobe.

INTRODUCTION: The aim of this study was to clarify the impact of middle hepatic artery reconstruction on the outcomes of duct-to-duct biliary anastomosis after living donor liver transplantation (LDLT) using the left lobe. MATERIALS AND METHODS: Among 258 patients who underwent LDLT using the left lobe, 216 patients who underwent hepatic artery reconstruction and one hepatic duct reconstruction with duct-to-duct interrupted anastomosis were divided into three groups: Group A (n = 123), one arterial stump with left hepatic artery reconstruction; Group B (n = 32), two arterial stumps with only left hepatic artery reconstruction; and Group C (n = 61), two arterial stumps with reconstruction of the left and middle hepatic arteries. The outcomes after LDLT were compared among the three groups. RESULTS: No hepatic artery complications occurred. Group B had a significantly greater incidence of anastomotic biliary stricture than Group C. A multivariate analysis with Cox regression revealed that being in Group B was the only significant independent risk factor for postoperative anastomotic biliary stricture after LDLT. CONCLUSIONS: Middle and left hepatic artery reconstruction is safe in LDLT and may prevent biliary stricture caused by dual hepatic artery reconstruction when the graft has left and middle hepatic artery stumps.

Clip on Staple Method Reduces Clinically Relevant Pancreatic Fistula After Distal Pancreatectomy.

BACKGROUND/AIM: In order to overcome postoperative pancreatic fistula (POPF) after distal pancreatectomy (DP), we have developed a new simple technique-Clip on Staple method. PATIENTS AND METHODS: In Clip on Staple method, pancreatic parenchyma was divided using a stapling device with a stepped-height staple design to make linear compression line, and thereafter, the full length of the staple line was reinforced by multiple clips. Clinical outcomes were retrospectively compared between Clip on Staple group (n=23) and Non-Clip group (n=38). RESULTS: The incidence of clinically relevant POPF (CR-POPF) was significantly lower in the Clip on Staple group than in the Non-Clip group (4.3 and 36.8%, p=0.005). Multivariate logistic regression analysis revealed that only Clip on Staple method was an independent predictive factor of a decrease in the occurrence of CR-POPF. CONCLUSION: The Clip on Staple method, a simple and easily applicable technique even in laparoscopic surgery, significantly reduced the occurrence of CR-POPF among patients undergoing DP.

End-to-End Learning Framework for IMU-Based 6-DOF Odometry.

This paper presents an end-to-end learning framework for performing 6-DOF odometry by using only inertial data obtained from a low-cost IMU. The proposed inertial odometry method allows leveraging inertial sensors that are widely available on mobile platforms for estimating their 3D trajectories. For this purpose, neural networks based on convolutional layers combined with a two-layer stacked bidirectional LSTM are explored from the following three aspects. First, two 6-DOF relative pose representations are investigated: one based on a vector in the spherical coordinate system, and the other based on both a translation vector and an unit quaternion. Second, the loss function in the network is designed with the combination of several 6-DOF pose distance metrics: mean squared error, translation mean absolute error, quaternion multiplicative error and quaternion inner product. Third, a multi-task learning framework is integrated to automatically balance the weights of multiple metrics. In the evaluation, qualitative and quantitative analyses were conducted with publicly-available inertial odometry datasets. The best combination of the relative pose representation and the loss function was the translation and quaternion together with the translation mean absolute error and quaternion multiplicative error, which obtained more accurate results with respect to state-of-the-art inertial odometry techniques.

Extensive Thrombectomy as a Legitimate Strategy in Living Donor Liver Transplantation With Advanced Portal Vein Thrombosis.

Management of portal vein thrombosis (PVT), especially advanced PVT involving the superior mesenteric vein (SMV), in living donor liver transplantation (LDLT) is challenging. There were 514 adults who underwent LDLT between 2005 and 2018 included in this retrospective study, and PVT was observed in 67 (13.0%) patients. The LDLT recipients with PVT were characterized by increased portal pressure at laparotomy (26.1 ± 6.0 versus 24.3 ± 5.9 mm Hg; P = 0.03) and at closure (16.8 ± 3.9 versus 15.6 ± 3.6 mm Hg; P = 0.02), increased operative blood loss (14.6 ± 29.7 versus 5.7 ± 6.3 L; P < 0.01), and decreased 1-year graft survival (83.5% versus 92.8%; P = 0.04). Among the 18 patients with atrophic or vanished portal vein on pre-LDLT computed tomography, significant portal atrophy was actually observed only in 1 (5.6%) patient during LDLT surgery. For advanced PVT (n = 7) involving SMV in era 1, we performed nonanatomical inflow reconstruction using interposition grafts, resulting in significant inflow problems in 4 (57.1%) patients. Thus, for the patients with advanced PVT (n = 4) in era 2, we abandoned nonanatomical reconstruction and applied extensive thrombectomy under ultrasound guidance with secure shunt ligation, resulting in no inflow problems and no graft loss. In conclusion, even for advanced PVT involving SMV, extensive thrombectomy under sonogram guidance followed by anatomical inflow reconstruction and shunt ligation is a legitimate strategy in adult LDLT with PVT.

Peritransplant kinetics of Mac-2-binding protein glycosylation isomer levels in living donor liver transplantation: its implication of posttransplant small-for-size syndrome.

BACKGROUND: Wisteria floribunda agglutinin positive human Mac-2 binding protein glycosylation isomer (M2BPGi) has recently developed as a noninvasive serum marker of liver fibrosis. Liver transplant candidates usually have high serum levels of M2BPGi due to advanced cirrhosis. The aim of the present study was to elucidate the kinetics of serum M2BPGi after liver transplantation and the relationships between the level of M2BPGi and graft function. METHODS: Fifteen recipients who underwent living donor liver transplantation (LDLT) between June 2015 and January 2016 and whose pretransplant, postoperative day (POD) 1, POD 3, and POD 7 sera were available for measuring M2BPGi were enrolled in this study. Small-for-size syndrome (SFSS) was defined as the presence of cholestasis (total bilirubin >10 mg/dL) on POD 7 and intractable ascites (>1 L/day on POD 14 or >500 ml/day on POD 28) without other specific causes. RESULTS: The median of pretransplant M2BPGi was 9.75 cutoff index (C.O.I.) (range, 3.04-24.49). There was neither any correlation between pretransplant M2BPGi and Model for End-Stage Liver Disease scores (r=0.416, P=0.123) nor Child-Turcotte-Pugh scores (r=-0.221, P=0.428). The levels of M2BPGi dramatically decreased after LDLT (median; 1.48 on POD 1, 1.47 on POD 3, 1.49 on POD 7). However, serum levels of M2BPGi rose again on POD 7 in some recipients and all 4 recipients with serum levels of M2BPGi exceeding 3.00 C.O.I. succumbed to SFSS later. When the cutoff of M2BPGi on POD 7 for predicting SFSS was determined to be 3.06 according to its receiver operating characteristic curve, both the sensitivity and the specificity for predicting later SFSS were 100%. CONCLUSIONS: The levels of M2BPGi dramatically decreased after LDLT. A re-rise of M2BPGi predicted later development of SFSS.

Solving Monocular Visual Odometry Scale Factor with Adaptive Step Length Estimates for Pedestrians Using Handheld Devices.

The urban environments represent challenging areas for handheld device pose estimation (i.e., 3D position and 3D orientation) in large displacements. It is even more challenging with low-cost sensors and computational resources that are available in pedestrian mobile devices (i.e., monocular camera and Inertial Measurement Unit). To address these challenges, we propose a continuous pose estimation based on monocular Visual Odometry. To solve the scale ambiguity and suppress the scale drift, an adaptive pedestrian step lengths estimation is used for the displacements on the horizontal plane. To complete the estimation, a handheld equipment height model, with respect to the Digital Terrain Model contained in Geographical Information Systems, is used for the displacement on the vertical axis. In addition, an accurate pose estimation based on the recognition of known objects is punctually used to correct the pose estimate and reset the monocular Visual Odometry. To validate the benefit of our framework, experimental data have been collected on a 0.7 km pedestrian path in an urban environment for various people. Thus, the proposed solution allows to achieve a positioning error of 1.6⁻7.5% of the walked distance, and confirms the benefit of the use of an adaptive step length compared to the use of a fixed-step length.

Incremental 3D Cuboid Modeling with Drift Compensation.

This paper presents a framework of incremental 3D cuboid modeling by using the mapping results of an RGB-D camera based simultaneous localization and mapping (SLAM) system. This framework is useful in accurately creating cuboid CAD models from a point cloud in an online manner. While performing the RGB-D SLAM, planes are incrementally reconstructed from a point cloud in each frame to create a plane map. Then, cuboids are detected in the plane map by analyzing the positional relationships between the planes, such as orthogonality, convexity, and proximity. Finally, the position, pose, and size of a cuboid are determined by computing the intersection of three perpendicular planes. To suppress the false detection of the cuboids, the cuboid shapes are incrementally updated with sequential measurements to check the uncertainty of the cuboids. In addition, the drift error of the SLAM is compensated by the registration of the cuboids. As an application of our framework, an augmented reality-based interactive cuboid modeling system was developed. In the evaluation at cluttered environments, the precision and recall of the cuboid detection were investigated, compared with a batch-based cuboid detection method, so that the advantages of our proposed method were clarified.