Development and Validation of an Artificial Intelligence Model for Detecting Rib Fractures on Chest Radiographs.

Background: Chest radiography is the standard method for detecting rib fractures. Our study aims to develop an artificial intelligence (AI) model that, with only a relatively small amount of training data, can identify rib fractures on chest radiographs and accurately mark their precise locations, thereby achieving a diagnostic accuracy comparable to that of medical professionals. Methods: For this retrospective study, we developed an AI model using 540 chest radiographs (270 normal and 270 with rib fractures) labeled for use with Detectron2 which incorporates a faster region-based convolutional neural network (R-CNN) enhanced with a feature pyramid network (FPN). The model's ability to classify radiographs and detect rib fractures was assessed. Furthermore, we compared the model's performance to that of 12 physicians, including six board-certified anesthesiologists and six residents, through an observer performance test. Results: Regarding the radiographic classification performance of the AI model, the sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) were 0.87, 0.83, and 0.89, respectively. In terms of rib fracture detection performance, the sensitivity, false-positive rate, and free-response receiver operating characteristic (JAFROC) figure of merit (FOM) were 0.62, 0.3, and 0.76, respectively. The AI model showed no statistically significant difference in the observer performance test compared to 11 of 12 and 10 of 12 physicians, respectively. Conclusions: We developed an AI model trained on a limited dataset that demonstrated a rib fracture classification and detection performance comparable to that of an experienced physician.

Effects of video-assisted preanesthetic patient education on preanesthetic interview time; A prospective, single-blinded, randomized controlled trial.

BACKGROUND: A preanesthetic evaluation interview with an anesthesiologist is essential for patient safety, however, it is not performed adequately owing to the excessive workload of doctors. This study aimed to determine whether video-assisted preanesthetic patient education can reduce patient interview time and solve the problem of excessive labor at a relatively low cost. METHODS: This study considered relatively healthy patients aged 19 to 65 years who were scheduled for elective surgery under general anesthesia. None of the patients had history of general anesthesia. Patients were randomly assigned 1:1 to Groups V and C. Group V watched the preanesthetic education video, while Group C did not. The duration of the preanesthetic evaluation interview was measured for all participants. The satisfaction of the anesthesiologist and patient with the preanesthetic evaluation procedure, anxiety of the patient, and vital signs during surgery were collected. RESULTS: A total of 33 patients in Group V watched the preanesthetic education video, while 31 patients in Group C did not. Group V spent significantly less time on the preanesthetic evaluation interview with an anesthesiologist than that of Group C (172.42 vs 196.68 seconds; P = .005). There was no difference in patient and anesthesiologist satisfaction between the 2 groups (P = .861 and P = .849, respectively). Patients' anxiety (P = .474), intraoperative mean blood pressure (P = .168), and heart rate (P = .934) did not differ between Groups V and C. CONCLUSION: Watching the informational video about anesthesia before preanesthetic evaluation could reduce the interview time by an average of 24 seconds, with no difference in patients' or doctors' satisfaction or anxiety compared to patients who did not watch it. Video-assisted preanesthetic patient education indicates that the load on anesthesiologists can be reduced.

Limited effect of sugammadex on postoperative pulmonary complications in patients undergoing spine surgery in prone position intraoperatively: A retrospective analysis of matched cohort data.

Postoperative pulmonary complications (PPCs) increase postoperative mortality, hospital stays, and healthcare costs. Whether the use of sugammadex could reduce PPCs remains controversial. This study aimed to determine if sugammadex could more effectively reduce PPCs than acetylcholinesterase inhibitor (AChEi) in patients who had undergone spine surgery, in prone position intraoperatively. From March 2019 to February 2021, adult patients who underwent elective spine surgery were eligible. Primary outcomes were PPCs (including atelectasis on chest radiograph, pneumonia, acute respiratory distress syndrome, and aspiration pneumonitis) and respiratory failure that occurred within 28 days after surgery. Secondary outcomes were length of hospital stay, in-hospital death, and readmission rate within 30 days. Patients were divided into 2 groups (Sugammadex group and AChEi group) and compared by 1:1 propensity score matching. Of a total of 823 patients who underwent spinal surgery, 627 were included. After 1:1 propensity matching, 142 patients were extracted for each group. PPCs occurred in 9 (6.3%) patients in both groups (P = 1.000). Respiratory failure occurred in 7 (4.9%) patients in the Sugammadex group and 5 (3.5%) patients in the AChEi group (P = .77). There was no significant difference in secondary outcomes between the 2 groups. Although there have been some evidences showing that the use of sugammadex can attenuate the development of PPCs, this study did not show positive effects of sugammadex on patients who underwent spine surgery in the prone position.

Comparison between pressure-controlled and manual ventilation during anesthetic induction in patients with expected difficult airway: A prospective randomized controlled trial.

BACKGROUND: Gastric insufflation can cause gastric regurgitation, which may be exacerbated in patients who are expected to have difficult airways. The purpose of this study was to investigate the difference in respiratory parameters and the frequency of gastric insufflation according to the ventilation mode during the anesthestic induction on patients who were predicted to have difficult facemask ventilation. METHODS: A total of eighty patients with expected airway difficulties were included. Patient were allocated to 2 groups (n = 40 each). In the manual ventilation group, ventilation was performed by putting a mask on the patient's face with 1-hand and adjusting the pressure limiting valve to 15 cm H2O. In the pressure-controlled ventilation group, a mask was held in place using 2-handed jaw-thrust maneuver. The pressure-controlled ventilation was applied and peak inspiration pressure was adjusted to achieve a tidal volume of 6 to 8 mL/kg. The primary outcome was the difference of the peak airway pressure between 2 groups every 30 seconds for 120 seconds duration of mask ventilation. We also evaluated respiratory variables including peak airway pressure, End-tidal carbon dioxide and also gastric insufflation using ultrasonography. RESULTS: The pressure-controlled ventilation group demonstrated lower peak airway pressure than the manual ventilation group (P = .005). End-tidal carbon dioxide was higher in the pressure-controlled ventilation group (P = .012). The incidence of gastric insufflation assessed by real-time ultrasonography of the gastric antrum was higher in the manual ventilation group than in the pressure-controlled ventilation group [3 (7.5%) vs 17 (42.5%), risk ratio (95% confidence interval): 0.06 to 0.56, P = .003]. CONCLUSIONS: Pressure-controlled ventilation during facemask ventilation in patients who were expected to have difficult airways showed a lower gastric insufflation rate with low peak airway pressure compared to manual ventilation.

Can Ensemble Deep Learning Identify People by Their Gait Using Data Collected from Multi-Modal Sensors in Their Insole?

Gait is a characteristic that has been utilized for identifying individuals. As human gait information is now able to be captured by several types of devices, many studies have proposed biometric identification methods using gait information. As research continues, the performance of this technology in terms of identification accuracy has been improved by gathering information from multi-modal sensors. However, in past studies, gait information was collected using ancillary devices while the identification accuracy was not high enough for biometric identification. In this study, we propose a deep learning-based biometric model to identify people by their gait information collected through a wearable device, namely an insole. The identification accuracy of the proposed model when utilizing multi-modal sensing is over 99%.

User Identification from Gait Analysis Using Multi-Modal Sensors in Smart Insole.

Recent studies indicate that individuals can be identified by their gait pattern. A number of sensors including vision, acceleration, and pressure have been used to capture humans' gait patterns, and a number of methods have been developed to recognize individuals from their gait pattern data. This study proposes a novel method of identifying individuals using null-space linear discriminant analysis on humans' gait pattern data. The gait pattern data consists of time series pressure and acceleration data measured from multi-modal sensors in a smart insole used while walking. We compare the identification accuracies from three sensing modalities, which are acceleration, pressure, and both in combination. Experimental results show that the proposed multi-modal features identify 14 participants with high accuracy over 95% from their gait pattern data of walking.

Local affine image matching and synthesis based on structural patterns.

A general purpose block-to-block affine transformation estimator is described. The estimator is based on Fourier slice analysis and Fourier spectral alignment. It shows encouraging performance in terms of both speed and accuracy compared to existing methods. The key elements of its success are attributed to the ability to: 1) locate an arbitrary number of affine invariant points in the spectrum that latch onto significant structural features; 2) match the estimated invariant points with the target spectrum by the slicewise phase-correlation; and 3) use affine invariant points to directly compute all linear parameters of the full affine transform by spectral alignment. Experimental results using a wide range of textures are presented. Potential applications include affine invariant image segmentation, registration, affine symmetric image coding, and motion analysis.

An affine symmetric image model and its applications.

Natural images contain considerable redundancy, some of which is successfully captured using recently developed directional wavelets. In this paper, an affine symmetric image model is considered. It provides a flexible scheme to exploit geometric redundancy. A patch of texture from an image is rotated, scaled and sheared to approximate other similar parts in the image, revealing the self-similarity relation. The general scheme is derived as follows. A texture model is required that identifies structural patterns. Then the affine symmetry is exploited between structural textures at a local level, the objective being to find the minimum residual error by estimating the affine transform relating two patches of texture. Having developed a local model, the methodology is extended to the whole image to estimate the global affine relation. This global model is further developed in a multiresolution framework for multiscale analysis, by which the self similarity of the image is exploited across space and scale. The multiresolution model can be applied to a series of practical problems. Experimental evaluation demonstrates the effectiveness of the approach in affine invariant texture segmentation and image approximation.

Decreases in CD8+ T, naive (CD4+CD45RA+) T, and B (CD19+) lymphocytes by exposure to manganese fume.

To examine the effects of exposure to manganese (Mn) on the cellular and humoral immune system in men, T lymphocyte subpopulations, B (CD19+) lymphocytes, natural killer (NK) cells, and serum immunoglobulins (i.e., IgG, IgA and IgM) together with total T (CD3+) lymphocytes and total lymphocytes were measured in blood samples from 21 welders mainly exposed to Mn fume with blood Mn (BMn) concentrations of 0.6-2.3 (mean 1.4) microg/dl and 21 healthy controls working in the same factory (BMn concentrations: 0.7 to 1.7, mean 1.1 microg/dl). The workers engaged in welding for 6 to 36 (mean 17) yr. All the study subjects were divided into 3 equally sized groups (n=14 for each group) according to BMn concentrations. Numbers of CD8+ T, total T (CD3+), B (CD19+), and total lymphocytes were significantly lower in high-BMn group than those in low-BMn group; the numbers of CD8+ T lymphocytes were significantly lower in moderate-BMn group compared to low-BMn group. After adjusting for age and smoking, significant inverse correlations between BMn concentrations and CD4+CD45RA+ T, CD4+ T, CD8+ T, CD3+ T, and total lymphocytes were found. We conclude that T lymphocytes, especially CD8+ and CD4+CD45RA+ T lymphocytes, as well as CD19+ B lymphocytes are affected by exposure to Mn fume.

DNA Damage of Lymphocytes in Volunteers after 4 hours Use of Mobile Phone.

OBJECTIVES: There has been gradually increasing concern about the adverse health effects of electromagnetic radiation originating from cell phones which are widely used in modern life. Cell phone radiation may affect human health by increasing free radicals of human blood cells. This study has been designed to identify DNA damage of blood cells by electromagnetic radiation caused by cell phone use. METHODS: This study investigated the health effect of acute exposure to commercially available cell phones on certain parameters such as an indicator of DNA damage for 14 healthy adult volunteers. Each volunteer during the experiment talked over the cell phone with the keypad facing the right side of the face for 4 hours. The single cell gel electrophoresis assay (Comet assay), which is very sensitive in detecting the presence of DNA strand-breaks and alkali-labile damage in individual cells, was used to assess peripheral blood cells (T-cells, B-cells, granulocytes) from volunteers before and after exposure to cell phone radiation. The parameters of Comet assay measured were Olive Tail Moment and Tail DNA %. RESULTS: The Olive Tail Moment of B-cells and granulocytes and Tail DNA % of B-cells and granulocytes were increased by a statistically significant extent after 4- hour use of a cell phone compared with controls. CONCLUSIONS: It is concluded that cell phone radiation caused the DNA damage during the 4 hours of experimental condition. Nonetheless, this study suggested that cell phone use may increase DNA damage by electromagnetic radiation and other contributing factors.

Application status and its affecting factors of double standard for multinational corporations in Korea.

OBJECTIVE: We intended to evaluate the double standard status and to identify factors of determining double standard criteria in multinational corporations of Korea, and specifically those in the occupational health and safety area. METHODS: A postal questionnaire had been sent, between August 2002 and September 2002, to multinational corporations in Korea. A double standard company was defined as those who answered in more than one item as adopting a different standard among the five items regarding double standard identification. By comparing double standard companies with equivalent standard companies, determinants for double standards were then identified using logistic regression analysis. RESULTS: Of multinational corporations, 45.1% had adopted a double standard. Based on the question naire's scale level, the factor of 'characteristic and size of multinational corporation' was found to have the most potent impact on increasing double standard risk. On the variable level, factors of 'number of affiliated companies' and 'existence of an auditing system with the parent company' showed a strong negative impact on double standard risk. CONCLUSIONS: Our study suggests that a distinctive approach is needed to manage the occupational safety and health for multinational corporations. This approach should be focused on the specific level of a corporation, not on a country level.

Increase of olfactory threshold in plating factory workers exposed to chromium in Korea.

To disclose the effects of chromium (Cr) on olfactory function, olfactory threshold tests were conducted on 27 male plating workers (Cr workers) with signs and symptoms of olfactory irritation but without nasal septum perforation or ulcer and on 34 male control subjects in Korean plating factories. The Cr workers had been exposed to Cr fume for 0.9 to 18.2 (mean 7.9) years; their blood Cr concentrations (0.16-3.69, mean 1.29 microg/dl) were significantly higher than those of the 34 control subjects (0.04-1.95, mean 0.55 microg/dl). Scores on recognition thresholds among the Cr workers were significantly higher than those of the control subjects (p < 0.05) and related positively and significantly to the exposure periods of the 27 Cr workers (p < 0.05). Olfactory thresholds were not significantly different between the Cr workers with and without nasal signs or symptoms, except that the scores on the recognition threshold were significantly higher in those experiencing difficulty with smell (p < 0.05). It is suggested that olfactory threshold is affected by Cr without development of nasal septum perforation or ulceration.