Towards safer imaging: A comparative study of deep learning-based denoising and iterative reconstruction in intraindividual low-dose CT scans using an in-vivo large animal model.

PURPOSE: Computed tomography (CT) scans are a significant source of medically induced radiation exposure. Novel deep learning-based denoising (DLD) algorithms have been shown to enable diagnostic image quality at lower radiation doses than iterative reconstruction (IR) methods. However, most comparative studies employ low-dose simulations due to ethical constraints. We used real intraindividual animal scans to investigate the dose-reduction capabilities of a DLD algorithm in comparison to IR. MATERIALS AND METHODS: Fourteen veterinarian-sedated alive pigs underwent 2 CT scans on the same 3rd generation dual-source scanner with two months between each scan. Four additional scans ensued each time, with mAs reduced to 50 %, 25 %, 10 %, and 5 %. All scans were reconstructed ADMIRE levels 2 (IR2) and a novel DLD algorithm, resulting in 280 datasets. Objective image quality (CT numbers stability, noise, and contrast-to-noise ratio) was measured via consistent regions of interest. Three radiologists independently rated all possible dataset combinations per time point for subjective image quality (-1 = inferior, 0 = equal, 1 = superior). The points were averaged for a semiquantitative score, and inter-rater agreement was measured using Spearman's correlation coefficient and adequately corrected mixed-effects modeling analyzed objective and subjective image quality. RESULTS: Neither dose-reduction nor reconstruction method negatively impacted CT number stability (p > 0.999). In objective image quality assessment, the lowest radiation dose achievable by DLD when comparing noise (p = 0.544) and CNR (p = 0.115) to 100 % IR2 was 25 %. Overall, inter-rater agreement of the subjective image quality ratings was strong (r ≥ 0.69, mean 0.93 ± 0.05, 95 % CI 0.92-0.94; each p < 0.001), and subjective assessments corroborated that DLD at 25 % radiation dose was comparable to 100 % IR2 in image quality, sharpness, and contrast (p ≥ 0.281). CONCLUSIONS: The DLD algorithm can achieve image quality comparable to the standard IR method but with a significant dose reduction of up to 75%. This suggests a promising avenue for lowering patient radiation exposure without sacrificing diagnostic quality.

How Real Are Computed Tomography Low Dose Simulations? An Investigational In-Vivo Large Animal Study.

OBJECTIVES: CT low-dose simulation methods have gained significant traction in protocol development, as they lack the risk of increased patient exposure. However, in-vivo validations of low-dose simulations are as uncommon as prospective low-dose image acquisition itself. Therefore, we investigated the extent to which simulated low-dose CT datasets resemble their real-dose counterparts. MATERIALS AND METHODS: Fourteen veterinarian-sedated alive pigs underwent three CT scans on the same third generation dual-source scanner with 2 months between each scan. At each time, three additional scans ensued, with mAs reduced to 50%, 25%, and 10%. All scans were reconstructed using wFBP and ADMIRE levels 1-5. Matching low-dose datasets were generated from the 100% scans using reconstruction-based and DICOM-based simulations. Objective image quality (CT numbers stability, noise, and signal-to-noise ratio) was measured via consistent regions of interest. Three radiologists independently rated all possible dataset combinations per time point for subjective image quality (-1=inferior, 0=equal, 1=superior). The points were averaged for a semiquantitative score, and inter-rater-agreement was measured using Spearman's correlation coefficient. A structural similarity index (SSIM) analyzed the voxel-wise similarity of the volumes. Adequately corrected mixed-effects analysis compared objective and subjective image quality. Multiple linear regression with three-way interactions measured the contribution of dose, reconstruction mode, simulation method, and rater to subjective image quality. RESULTS: There were no significant differences between objective and subjective image quality of reconstruction-based and DICOM-based simulation on all dose levels (p≥0.137). However, both simulation methods produced significantly lower objective image quality than real-dose images below 25% mAs due to noise overestimation (p<0.001; SSIM≤89±3). Overall, inter-rater-agreement was strong (r≥0.68, mean 0.93±0.05, 95% CI 0.92-0.94; each p<0.001). In regression analysis, significant decreases in subjective image quality were observed for lower radiation doses (b ≤ -0.387, 95%CI -0.399 to -0.358; p<0.001) but not for reconstruction modes, simulation methods, raters, or three-way interactions (p≥0.103). CONCLUSION: Simulated low-dose CT datasets are subjectively and objectively indistinguishable from their real-dose counterparts down to 25% mAs, making them an invaluable tool for efficient low-dose protocol development.

Role of Rendezvous-Procedure in the Treatment of Complications after Laparoscopic Sleeve Gastrectomy.

INTRODUCTION: Laparoscopic sleeve gastrectomy is one of the most commonly performed bariatric procedures worldwide with good results, high patient acceptance, and low complication rates. The most relevant perioperative complication is the staple line leak. For the treatment of this complication, endoscopic negative pressure therapy has proven particularly effective. The correct time to start endoscopic negative pressure therapy has not been the subject of studies to date. METHODS: Twelve patients were included in this retrospective data analysis over three years. Endoscopic negative pressure therapy was carried out using innovative open pore suction devices. Patients were treated with simultaneous surgery and endoscopy, so called rendezvous-procedure (Group A) or solely endoscopically, or in sequence surgically and endoscopically (Group B). Therapy data of the procedures and outcome measures, including duration of therapy, therapy success, and change of treatment strategy, were collected and analysed. RESULTS: In each group, six patients were treated (mean age 52.96 years, 4 males, 8 females). Poor initial clinical situation, time span of endoscopic negative pressure therapy (Group A 31 days vs. Group B 18 days), and mean length of hospital stay (Group A 39.5 days vs. Group B 20.17 days) were higher in patients with rendezvous procedures. One patient in Group B died during the observation time. DISCUSSION: Rendezvous procedures for patients with staple line leaks after sleeve gastrectomy is indicated for serious ill patients with perigastric abscesses and in need of laparoscopic lavage. The one-stage complication management with the rendezvous procedure seems not to result in an obvious advantage in the further outcome in patients with staple line leaks after laparoscopic sleeve gastrectomy.

Endoscopic Vacuum Therapy for Staple Line Leaks after Sleeve Gastrectomy.

PURPOSE: Management of staple line leaks (SLL) after sleeve gastrectomy (SG) is challenging. The aim of this study was to evaluate the effectiveness of a novel endoscopic vacuum therapy (EVT) modality in the management of sleeve leaks. MATERIALS AND METHODS: Eight patients were treated with EVT for SLL. Therapy data and outcome measures including duration of therapy, therapy success, and change of treatment strategy were collected and analyzed. RESULTS: During the study period, SLL occurred in 1.6% of patients who underwent SG. After 9.8 ± 8.6 days of EVT, 3.3 ± 2.2 endoscopies, and 19 ± 15.1 days of hospitalization, endoscopic treatment using EVT was successful in seven out of eight patients (87.5%). CONCLUSIONS: EVT is an effective method for the management of staple line leaks after sleeve gastrectomy. The use of the intraluminal open-pore film drainage (OFD) could be considered as an advantageous modality of EVT, regarding placement and complications.

Automated closed-loop management of body temperature using forced-air blankets: preliminary feasibility study in a porcine model.

BACKGROUND: Management of a patient's body temperature is an important aspect of care that should be addressed by targeted temperature management (TTM). Often, non-invasive methods like forced-air blankets are used. Especially in the operating room this management may be a subsidiary and repetitive task requiring constant observation of the patient's body temperature and adaption using the limited set of available settings. Thus, automation of TTM is a feasible target to improve patient outcome and reduce caregiver workload. METHODS: A Philips IntelliVue MP 50 patient monitor with an arterial PiCCO catheter system was used to measure patient blood temperature. Thermal management was performed with a 3M Bair Hugger 755 warming unit with forced air blankets. The warming unit was extended by a computer interface to allow for remote and automated control. A proposed closed-loop algorithm reads the measured temperature and performs automated control of the 3M Bair Hugger. Evaluation was performed in an experimental intensive care setting for animal studies. Two fully automated trials are compared with two manual and two uncontrolled trials in the same study setting using six female pigs for prolonged observation times of up to 90 hours in each trial. RESULTS: The developed system and proposed algorithm allow more precise temperature management by keeping a set target temperature within a range of ± 0.5 °C in 88% of the observation time and within a range of ± 1.0 °C at all times. The proposed algorithm yielded better performance than did manual control or uncontrolled trials. It was able to adapt to individual patient needs as it is more dynamic than look-up table approaches with fixed settings for various temperatures. CONCLUSIONS: Closed-loop TTM using non-invasive forced-air warming blankets was successfully tested in a porcine study with the proposed hardware interface and control algorithm. This automation can be beneficial for patient outcome and can reduce caregiver workload and patient risk in clinical settings. As temperature readings are most often available, existing devices like the 3M Bair Hugger can easily be expanded. However, even if clinical application is feasible, open questions regarding approval and certification of such automated systems within the current legal situation still need to be answered.

Fully automated life support: an implementation and feasibility pilot study in healthy pigs.

BACKGROUND: Automated systems are available in various application areas all over the world for the purpose of reducing workload and increasing safety. However, such support systems that would aid caregivers are still lacking in the medical sector. With respect to workload and safety, especially, the intensive care unit appears to be an important and challenging application field. Whereas many closed-loop subsystems for single applications already exist, no comprehensive system covering multiple therapeutic aspects and interactions is available yet. This paper describes a fully closed-loop intensive care therapy and presents a feasibility analysis performed in three healthy pigs over a period of 72 h each to demonstrate the technical and practical implementation of automated intensive care therapy. METHODS: The study was performed in three healthy, female German Landrace pigs under general anesthesia with endotracheal intubation. An arterial and a central venous line were implemented, and a suprapubic urinary catheter was inserted. Electrolytes, glucose levels, acid-base balance, and respiratory management were completely controlled by an automated fuzzy logic system based on individual targets. Fluid management by adaption of the respective infusion rates for the individual parameters was included. RESULTS: During the study, no manual modification of the device settings was allowed or required. Homoeostasis in all animals was kept stable during the entire observation period. All remote-controlled parameters were maintained within physiological ranges for most of the time (free arterial calcium 73%, glucose 98%, arterial base excess 89%, and etCO2 98%). Subsystem interaction was analyzed. CONCLUSIONS: In the presented study, we demonstrate the feasibility of a fully closed-loop system, for which we collected high-resolution data on the interaction and response of the different subsystems. Further studies should use big data approaches to analyze and investigate the interactions between the subsystems in more detail.

Pathophysiological central nervous system changes in a porcine model of acetaminophen-induced acute liver failure.

BACKGROUND: Critical care management of patients suffering from acute liver failure (ALF) continues to be challenging. Animal models studying the pathophysiological central nervous system alterations during the course of ALF provide an opportunity to improve diagnostic and therapeutic strategies. The aim of this study was to analyse the course of cerebral oxygenation in addition to conventional neuromonitoring during the course of acetaminophen-induced ALF. METHODS: ALF was induced by intrajejunal acetaminophen administration in 20 German landrace pigs. All animals underwent invasive hemodynamic and neuromonitoring and were maintained under standardized intensive care support. Neuromonitoring consisted of continuous intraparenchymatous recording of intracranial pressure and brain partial oxygen pressure. Hemodynamic and ventilation parameters were continuously recorded; laboratory parameters were analysed every eight hours. Mean values were compared using the Wilcoxon test. RESULTS: Acute liver failure occurred in all intoxicated animals after 23±2h, resulting in death due to ALF after further 15±2h. Continuous neuromonitoring was performed in all animals during the whole experiment without observing signs of intracranial haemorrhage. Two hours after manifestation of ALF an increase in brain tissue oxygen (PtiO2) was observed. Brain oxygenation stayed stable until nine hours before death. Intracranial pressure (ICP) remained basically at a plateau level until manifestation of ALF. In the following ten hours a linear and slow increase was observed until five hours before death, followed by a fast and continuous rise in ICP to a final level of 35±1mmHg. Cerebral perfusion pressure (CPP) began to decrease 25h prior to exitus, further decreasing to 18±2mmHg at the end of the experiment. A strong negative linear correlation was found between PtiO2 and ICP (R=0.97). Arterial partial pressure of oxygen (PaO2) below 100mmHg was associated with lower PtiO2 levels. Changes in arterial partial pressure of carbon dioxide (PaC02) did not influence PtiO2 values. Hemoglobin values below 7g/dl were associated with lower PtiO2 values. CONCLUSIONS: The results of our experiments demonstrate that ICP and PtiO2 measurements indicate impending damage well before serious complications occur and their use should be considered in order to protect endangered brain function in the presence of acetaminophen-induced ALF.

Algorithm-based arterial blood sampling recognition increasing safety in point-of-care diagnostics.

AIM: To detect blood withdrawal for patients with arterial blood pressure monitoring to increase patient safety and provide better sample dating. METHODS: Blood pressure information obtained from a patient monitor was fed as a real-time data stream to an experimental medical framework. This framework was connected to an analytical application which observes changes in systolic, diastolic and mean pressure to determine anomalies in the continuous data stream. Detection was based on an increased mean blood pressure caused by the closing of the withdrawal three-way tap and an absence of systolic and diastolic measurements during this manipulation. For evaluation of the proposed algorithm, measured data from animal studies in healthy pigs were used. RESULTS: Using this novel approach for processing real-time measurement data of arterial pressure monitoring, the exact time of blood withdrawal could be successfully detected retrospectively and in real-time. The algorithm was able to detect 422 of 434 (97%) blood withdrawals for blood gas analysis in the retrospective analysis of 7 study trials. Additionally, 64 sampling events for other procedures like laboratory and activated clotting time analyses were detected. The proposed algorithm achieved a sensitivity of 0.97, a precision of 0.96 and an F1 score of 0.97. CONCLUSION: Arterial blood pressure monitoring data can be used to perform an accurate identification of individual blood samplings in order to reduce sample mix-ups and thereby increase patient safety.

Porcine model characterizing various parameters assessing the outcome after acetaminophen intoxication induced acute liver failure.

AIM: To investigate the changes of hemodynamic and laboratory parameters during the course of acute liver failure following acetaminophen overdose. METHODS: Eight pigs underwent a midline laparotomy following jejunal catheter placement for further acetaminophen intoxication and positioning of a portal vein Doppler flow-probe. Acute liver failure was realized by intrajejunal acetaminophen administration in six animals, two animals were sham operated. All animals were invasively monitored and received standardized intensive care support throughout the study. Portal blood flow, hemodynamic and ventilation parameters were continuously recorded. Laboratory parameters were analysed every eight hours. Liver biopsies were sampled every 24 h following intoxication and upon autopsy. RESULTS: Acute liver failure (ALF) occurred after 28 ± 5 h resulted in multiple organ failure and death despite maximal support after further 21 ± 1 h (study end). Portal blood flow (baseline 1100 ± 156 mL/min) increased to a maximum flow of 1873 ± 175 mL/min at manifestation of ALF, which was significantly elevated (P < 0.01). Immediately after peaking, portal flow declined rapidly to 283 ± 135 mL/min at study end. Thrombocyte values (baseline 307 × 103/µL ± 34 × 103/µL) of intoxicated animals declined slowly to values of 145 × 103/µL ± 46 × 103/µL when liver failure occurred. Subsequent appearance of severe thrombocytopenia in liver failure resulted in values of 11 × 103/µL ± 3 × 103/µL preceding fatality within few hours which was significant (P > 0.01). CONCLUSION: Declining portal blood flow and subsequent severe thrombocytopenia after acetaminophen intoxication precede fatality in a porcine acute liver failure model.

Correlation of the intracranial pressure to the central venous pressure in the late phase of acute liver failure in a porcine model.

Volume loading is a common method used to ensure adequate circulation. However, in the late phase of acute liver failure complications that often lead to death are cerebral swelling and brainstem edema, which are considered to result from increasing intracranial pressure (ICP). In former studies cerebral venous pressure (CVP) and ICP were reported to be independent entities. Acute liver failure was induced in 25 German land race pigs by acetaminophen intoxication. CVP and ICP were measured continuously. Hydroxyethyl starch solution and noradrenalin were administered to stabilize the circulation at a mean arterial pressure above 60mmHg. There is an increasing correlation in quantity and quality between the CVP and ICP in the last 24 h before exitus. Beginning with a slope of 0.24 (ICP against CVP) and a low correlation coefficient of 0.08. 24h before exitus, this situation remained stable until 16 h to exitus (m = 0.22, r = 0.1). The correlation increased from 16 to 8 h prior to exitus to a slope of m = 0.5 and a correlation of r = 0.3 and remained until exitus. In late acute liver failure it seems therefore clinically reasonable to keep circulation within an adequate range by the use of noradrenalin and to avoid fluid overload.

The enterohepatic circulation of amanitin: kinetics and therapeutical implications.

BACKGROUND: Amatoxin poisoning induces a delayed onset of acute liver failure which might be explained by the prolonged persistence of the toxin in the enterohepatic circulation. Aim of the study was to demonstrate amanitin kinetics in the enterohepatic circulation. METHODS: Four pigs underwent α-amanitin intoxication receiving 0.35 mg/kg (n=2) or 0.15 mg/kg (n=2) intraportally. All pigs remained under general anesthesia throughout the observation period of 72 h. Laboratory values and amanitin concentration in systemic and portal plasma, bile and urine samples were measured. RESULTS: Amanitin concentrations measured 5h after intoxication of 219±5ng/mL (0.35 mg/kg) and 64±3 (0.15 mg/kg) in systemic plasma and 201±8ng/mL, 80±13ng/mL in portal plasma declined to baseline levels within 24h. Bile concentrations simultaneously recorded showed 153±28ng/mL and 99±58ng/mL and decreased slightly delayed to baseline within 32 h. No difference between portal and systemic amanitin concentration was detected after 24h. CONCLUSIONS: Amanitin disappeared almost completely from systemic and enterohepatic circulation within 24 h. Systemic detoxification and/or interrupting the enterohepatic circulation at a later date might be poorly effective.