Globally optimal OCT surface segmentation using a constrained IPM optimization.

Segmentation of multiple surfaces in optical coherence tomography (OCT) images is a challenging problem, further complicated by the frequent presence of weak boundaries, varying layer thicknesses, and mutual influence between adjacent surfaces. The traditional graph-based optimal surface segmentation method has proven its effectiveness with its ability to capture various surface priors in a uniform graph model. However, its efficacy heavily relies on handcrafted features that are used to define the surface cost for the "goodness" of a surface. Recently, deep learning (DL) is emerging as a powerful tool for medical image segmentation thanks to its superior feature learning capability. Unfortunately, due to the scarcity of training data in medical imaging, it is nontrivial for DL networks to implicitly learn the global structure of the target surfaces, including surface interactions. This study proposes to parameterize the surface cost functions in the graph model and leverage DL to learn those parameters. The multiple optimal surfaces are then simultaneously detected by minimizing the total surface cost while explicitly enforcing the mutual surface interaction constraints. The optimization problem is solved by the primal-dual interior-point method (IPM), which can be implemented by a layer of neural networks, enabling efficient end-to-end training of the whole network. Experiments on spectral-domain optical coherence tomography (SD-OCT) retinal layer segmentation demonstrated promising segmentation results with sub-pixel accuracy.

Deep learning from atrioventricular plane displacement in patients with Takotsubo syndrome: lighting up the black-box.

Aims: The spatiotemporal deep convolutional neural network (DCNN) helps reduce echocardiographic readers' erroneous 'judgement calls' on Takotsubo syndrome (TTS). The aim of this study was to improve the interpretability of the spatiotemporal DCNN to discover latent imaging features associated with causative TTS pathophysiology. Methods and results: We applied gradient-weighted class activation mapping analysis to visualize an established spatiotemporal DCNN based on the echocardiographic videos to differentiate TTS (150 patients) from anterior wall ST-segment elevation myocardial infarction (STEMI, 150 patients). Forty-eight human expert readers interpreted the same echocardiographic videos and prioritized the regions of interest on myocardium for the differentiation. Based on visualization results, we completed optical flow measurement, myocardial strain, and Doppler/tissue Doppler echocardiography studies to investigate regional myocardial temporal dynamics and diastology. While human readers' visualization predominantly focused on the apex of the heart in TTS patients, the DCNN temporal arm's saliency visualization was attentive on the base of the heart, particularly at the atrioventricular (AV) plane. Compared with STEMI patients, TTS patients consistently showed weaker peak longitudinal displacement (in pixels) in the basal inferoseptal (systolic: 2.15 ± 1.41 vs. 3.10 ± 1.66, P < 0.001; diastolic: 2.36 ± 1.71 vs. 2.97 ± 1.69, P = 0.004) and basal anterolateral (systolic: 2.70 ± 1.96 vs. 3.44 ± 2.13, P = 0.003; diastolic: 2.73 ± 1.70 vs. 3.45 ± 2.20, P = 0.002) segments, and worse longitudinal myocardial strain in the basal inferoseptal (-8.5 ± 3.8% vs. -9.9 ± 4.1%, P = 0.013) and basal anterolateral (-8.6 ± 4.2% vs. -10.4 ± 4.1%, P = 0.006) segments. Meanwhile, TTS patients showed worse diastolic mechanics than STEMI patients (E'/septal: 5.1 ± 1.2 cm/s vs. 6.3 ± 1.5 cm/s, P < 0.001; S'/septal: 5.8 ± 1.3 cm/s vs. 6.8 ± 1.4 cm/s, P < 0.001; E'/lateral: 6.0 ± 1.4 cm/s vs. 7.9 ± 1.6 cm/s, P < 0.001; S'/lateral: 6.3 ± 1.4 cm/s vs. 7.3 ± 1.5 cm/s, P < 0.001; E/E': 15.5 ± 5.6 vs. 12.5 ± 3.5, P < 0.001). Conclusion: The spatiotemporal DCNN saliency visualization helps identify the pattern of myocardial temporal dynamics and navigates the quantification of regional myocardial mechanics. Reduced AV plane displacement in TTS patients likely correlates with impaired diastolic mechanics.

Segmentation quality assessment by automated detection of erroneous surface regions in medical images.

Despite the advancement in deep learning-based semantic segmentation methods, which have achieved accuracy levels of field experts in many computer vision applications, the same general approaches may frequently fail in 3D medical image segmentation due to complex tissue structures, noisy acquisition, disease-related pathologies, as well as the lack of sufficiently large datasets with associated annotations. For expeditious diagnosis and quantitative image analysis in large-scale clinical trials, there is a compelling need to predict segmentation quality without ground truth. In this paper, we propose a deep learning framework to locate erroneous regions on the boundary surfaces of segmented objects for quality control and assessment of segmentation. A Convolutional Neural Network (CNN) is explored to learn the boundary related image features of multi-objects that can be used to identify location-specific inaccurate segmentation. The predicted error locations can facilitate efficient user interaction for interactive image segmentation (IIS). We evaluated the proposed method on two data sets: Osteoarthritis Initiative (OAI) 3D knee MRI and 3D calf muscle MRI. The average sensitivity scores of 0.95 and 0.92, and the average positive predictive values of 0.78 and 0.91 were achieved, respectively, for erroneous surface region detection of knee cartilage segmentation and calf muscle segmentation. Our experiment demonstrated promising performance of the proposed method for segmentation quality assessment by automated detection of erroneous surface regions in medical images.

Patch-wise 3D segmentation quality assessment combining reconstruction and regression networks.

Purpose: General deep-learning (DL)-based semantic segmentation methods with expert level accuracy may fail in 3D medical image segmentation due to complex tissue structures, lack of large datasets with ground truth, etc. For expeditious diagnosis, there is a compelling need to predict segmentation quality without ground truth. In some medical imaging applications, maintaining the quality of segmentation is crucial to the localized regions where disease is prevalent rather than just globally maintaining high-average segmentation quality. We propose a DL framework to identify regions of segmentation inaccuracies by combining a 3D generative adversarial network (GAN) and a convolutional regression network. Approach: Our approach is methodologically based on the learned ability to reconstruct the original images identifying the regions of location-specific segmentation failures, in which the reconstruction does not match the underlying original image. We use conditional GAN to reconstruct input images masked by the segmentation results. The regression network is trained to predict the patch-wise Dice similarity coefficient (DSC), conditioned on the segmentation results. The method relies directly on the extracted segmentation related features and does not need to use ground truth during the inference phase to identify erroneous regions in the computed segmentation. Results: We evaluated the proposed method on two public datasets: osteoarthritis initiative 4D (3D + time) knee MRI (knee-MR) and 3D non-small cell lung cancer CT (lung-CT). For the patch-wise DSC prediction, we observed the mean absolute errors of 0.01 and 0.04 with the independent standard for the knee-MR and lung-CT data, respectively. Conclusions: This method shows promising results in localizing the erroneous segmentation regions that may aid the downstream analysis of disease diagnosis and prognosis prediction.

KCB-Net: A 3D knee cartilage and bone segmentation network via sparse annotation.

Knee cartilage and bone segmentation is critical for physicians to analyze and diagnose articular damage and knee osteoarthritis (OA). Deep learning (DL) methods for medical image segmentation have largely outperformed traditional methods, but they often need large amounts of annotated data for model training, which is very costly and time-consuming for medical experts, especially on 3D images. In this paper, we report a new knee cartilage and bone segmentation framework, KCB-Net, for 3D MR images based on sparse annotation. KCB-Net selects a small subset of slices from 3D images for annotation, and seeks to bridge the performance gap between sparse annotation and full annotation. Specifically, it first identifies a subset of the most effective and representative slices with an unsupervised scheme; it then trains an ensemble model using the annotated slices; next, it self-trains the model using 3D images containing pseudo-labels generated by the ensemble method and improved by a bi-directional hierarchical earth mover's distance (bi-HEMD) algorithm; finally, it fine-tunes the segmentation results using the primal-dual Internal Point Method (IPM). Experiments on four 3D MR knee joint datasets (the SKI10 dataset, OAI ZIB dataset, Iowa dataset, and iMorphics dataset) show that our new framework outperforms state-of-the-art methods on full annotation, and yields high quality results for small annotation ratios even as low as 10%.

Predicting adverse cardiac events in sarcoidosis: deep learning from automated characterization of regional myocardial remodeling.

Recognizing early cardiac sarcoidosis (CS) imaging phenotypes can help identify opportunities for effective treatment before irreversible myocardial pathology occurs. We aimed to characterize regional CS myocardial remodeling features correlating with future adverse cardiac events by coupling automated image processing and data analysis on cardiac magnetic resonance (CMR) imaging datasets. A deep convolutional neural network (DCNN) was used to process a CMR database of a 10-year cohort of 117 consecutive biopsy-proven sarcoidosis patients. The maximum relevance - minimum redundancy method was used to select the best subset of all the features-24 (from manual processing) and 232 (from automated processing) left ventricular (LV) structural/functional features. Three machine learning (ML) algorithms, logistic regression (LogR), support vector machine (SVM) and multi-layer neural networks (MLP), were used to build classifiers to categorize endpoints. Over a median follow-up of 41.8 (inter-quartile range 20.4-60.5) months, 35 sarcoidosis patients experienced a total of 43 cardiac events. After manual processing, LV ejection fraction (LVEF), late gadolinium enhancement, abnormal segmental wall motion, LV mass (LVM), LVMI index (LVMI), septal wall thickness, lateral wall thickness, relative wall thickness, and wall thickness of 9 (out of 17) individual LV segments were significantly different between patients with and without endpoints. After automated processing, LVEF, end-diastolic volume, end-systolic volume, LV mass and wall thickness of 92 (out of 216) individual LV segments were significantly different between patients with and without endpoints. To achieve the best predictive performance, ML algorithms selected lateral wall thickness, abnormal segmental wall motion, septal wall thickness, and increased wall thickness of 3 individual segments after manual image processing, and selected end-diastolic volume and 7 individual segments after automated image processing. LogR, SVM and MLP based on automated image processing consistently showed better predictive accuracies than those based on manual image processing. Automated image processing with a DCNN improves data resolution and regional CS myocardial remodeling pattern recognition, suggesting that a framework coupling automated image processing with data analysis can help clinical risk stratification.

Spatio-temporal hybrid neural networks reduce erroneous human "judgement calls" in the diagnosis of Takotsubo syndrome.

BACKGROUND: We investigate whether deep learning (DL) neural networks can reduce erroneous human "judgment calls" on bedside echocardiograms and help distinguish Takotsubo syndrome (TTS) from anterior wall ST segment elevation myocardial infarction (STEMI). METHODS: We developed a single-channel (DCNN[2D SCI]), a multi-channel (DCNN[2D MCI]), and a 3-dimensional (DCNN[2D+t]) deep convolution neural network, and a recurrent neural network (RNN) based on 17,280 still-frame images and 540 videos from 2-dimensional echocardiograms in 10 years (1 January 2008 to 1 January 2018) retrospective cohort in University of Iowa (UI) and eight other medical centers. Echocardiograms from 450 UI patients were randomly divided into training and testing sets for internal training, testing, and model construction. Echocardiograms of 90 patients from the other medical centers were used for external validation to evaluate the model generalizability. A total of 49 board-certified human readers performed human-side classification on the same echocardiography dataset to compare the diagnostic performance and help data visualization. FINDINGS: The DCNN (2D SCI), DCNN (2D MCI), DCNN(2D+t), and RNN models established based on UI dataset for TTS versus STEMI prediction showed mean diagnostic accuracy 73%, 75%, 80%, and 75% respectively, and mean diagnostic accuracy of 74%, 74%, 77%, and 73%, respectively, on the external validation. DCNN(2D+t) (area under the curve [AUC] 0·787 vs. 0·699, P = 0·015) and RNN models (AUC 0·774 vs. 0·699, P = 0·033) outperformed human readers in differentiating TTS and STEMI by reducing human erroneous judgement calls on TTS. INTERPRETATION: Spatio-temporal hybrid DL neural networks reduce erroneous human "judgement calls" in distinguishing TTS from anterior wall STEMI based on bedside echocardiographic videos. FUNDING: University of Iowa Obermann Center for Advanced Studies Interdisciplinary Research Grant, and Institute for Clinical and Translational Science Grant. National Institutes of Health Award (1R01EB025018-01).

BK polyomavirus in renal transplants: role of electron microscopy and immunostaining in detecting early infection.

Reactivation of BK polyomavirus (BKV) is increasingly recognized as a cause of failure of renal allografts. Since no specific treatment is available for this infection, early diagnosis is important, as it allows for early intervention and possible recovery of renal function. Forty-four consecutive renal transplant biopsies performed over a 2-year period were included in the study. In addition to evaluation of renal biopsy tissue sections using routine histochemical stains, CD3, CD20, BK virus immunostains using the specific BK virus and the SV40 antibodies and electron microscopy studies were performed. None of the transplant cases but one exhibited classical histologic viral changes. Viral particles were seen by EM in 19%, and BK-virus positivity was identified in only 43% of these cases. CD20-rich inflammatory infiltrates predominated in cases in which either positive BK stain and/or viral particles were identified ultrastructurally. A combined approach using electron microscopic and immunohistochemical evaluation can be utilized effectively to identify BK virus-associated nephropathy at an early phase facilitating early clinical intervention.

Pretransplant bilateral hand-assisted laparoscopic nephrectomy in adult patients with polycystic kidney disease.

Laparoscopic procedures continue to gain popularity over traditional open procedures for a number of abdominal and pelvic surgeries. With increasing experience, the application of this technique is rising because it provides an alternative, less invasive, approach to various surgical procedures. Herein, we report our experience with adult patients with polycystic kidney disease, requiring bilateral laparoscopic nephrectomy before renal transplantation.

Isopropyl alcohol intoxication: a diagnostic challenge.

Isopropyl alcohol (IPA) is an ingredient of commonly used household solutions. Accidental and suicidal ingestion of IPA sometimes can be fatal if it goes unrecognized and untreated. There are few published reports on IPA intoxication. We describe a case of repeated IPA ingestion in a single individual, followed by a review of the literature on the subject. The differential diagnosis, diagnostic pitfalls, and therapeutic interventions in patients with IPA intoxications are discussed.

Salvage of poorly developed arteriovenous fistulae with percutaneous ligation of accessory veins.

Many arteriovenous (AV) fistulae fail to achieve an adequate blood flow or size for successful cannulation because of accessory veins. We describe a simple technique to ligate accessory veins that does not require a surgical incision. In this retrospective study, 17 end-stage renal disease patients underwent ligation of accessory veins of poorly developed AV fistulae. There were 14 men and 3 women, and their average age was 50 +/- 13 years. There were 14 radiocephalic and 3 brachiocephalic fistulae. After identifying accessory veins with a fistulogram, two nonabsorbable 2-0 polypropylene (Prolene) sutures were placed percutaneously around each accessory vein in proximity to the AV fistula. Successful ligation was confirmed with a repeat fistulogram. This procedure was undertaken after 4 +/- 3 months following surgical placement. Successful maturation was defined as adequate blood flow to support effective hemodialysis and adequate caliber to allow for repeated cannulation with a 15G or 16G needle. Of 17 AV fistulae, 15 (88%) successfully matured 1.7 +/- 1 month (range, 0.3 to 6 months) after the procedure. The average number of accessory veins ligated was 1.7 +/- 0.8 (range, 1 to 3). All AV fistulae that matured after ligation of accessory veins were functioning at 44.5 +/- 12 weeks after first use. A technique for salvaging nonmaturing AV fistulae not requiring surgical cutdown for ligation of accessory veins is described. AV fistulae mature quickly after ligation of accessory veins. This is a rapid and safe procedure that can increase the prevalence of AV fistulae.

Native quadruple-valve endocarditis caused by Enterococcus faecalis.

We report the case of a patient with postoperative, hospital-acquired, quadruple-valve endocarditis caused by Enterococcus faecalis on presumed normal native valves. During a cervical laminectomy, the patient had a non-ST-elevation myocardial infarction that was treated conservatively. In the intensive care unit, the patient became febrile and developed a new 2/6 systolic murmur. Blood cultures grew E. faecalis, and the patient was given antibiotics. Postoperative transthoracic echocardiography and transesophageal echocardiography revealed vegetations on all 4 heart valves. Subsequently, the patient was moved to another facility and died. No autopsy was performed. E. faecalis is the third-most-common cause of bacterial endocarditis overall; however, it is rarely found in multiple-valve, hospital-acquired endocarditis. Although transthoracic echocardiography is a powerful diagnostic tool, transesophageal echocardiography increases the sensitivity and specificity to about 90%. In our patient, the diagnosis of native quadruple-valve endocarditis would not have not been made without the use of transesophageal echocardiography.

Retrieval of a fractured piece of Tessio catheter with a snare using a transcutaneous transvenous approach.

Approximately 20% of end-stage renal disease patients requiring hemodialysis have central dialysis catheters as their vascular access. The major cause of central dialysis catheters removal or revision is infection or occlusion. Catheter occlusions may occur as a result of thrombosis or fibrin sheath formation. However, the presence of a fractured dialysis catheter tip requiring immediate extraction to prevent serious complications is rare. Herein we present the case of a central dialysis catheter referred to us for malfunction. An incidental finding was a piece of catheter that had broken off the venous port and was seen in the right atrium. The retrieval and subsequent placement of a new central dialysis catheter are outlined.

Fluoroscopy-assisted placement of peritoneal dialysis catheters by nephrologists.

In the early 1950s and 1960s, peritoneal dialysis (PD) was used primarily to treat patients with acute renal failure. Continuous ambulatory peritoneal dialysis (CAPD) was introduced in 1976 and continues to gain popularity as an effective method of renal replacement therapy for patients with end-stage renal disease (ESRD). The PD catheter is inserted into the abdominal cavity either by a surgeon, interventional radiologist, or nephrologist. We have adopted a percutaneous approach with fluoroscopic guidance for PD catheter insertion that is easy, safe, and provides good patency and infection rate results. In this article we describe the technique and our results. From August 2000 to May 2003, 34 PD catheters out of 36 were successfully inserted using the percutaneous fluoroscopic technique in selected patients referred from the nephrology clinic. All the PD catheters were placed in our Interventional Nephrology Vascular Suite by nephrologists.

Pancreatic transplantation: evaluation and management.

More than 2 million people in the United States have type 1 diabetes mellitus. Pancreatic transplantation has emerged as the single most effective means of achieving normal glucose homeostasis in this patient population. Newer immunosuppressive agents and surgical techniques continue to evolve, resulting in improved long-term graft and patient survival. Herein, an understanding of the evaluation, technical aspects, and perioperative management of pancreas transplantation is outlined.

Port catheter placement by nephrologists in an interventional nephrology training program.

We retrospectively reviewed all subcutaneous single- and double-lumen port catheters (PCs) inserted by interventional nephrologists at our institution to determine the success rate, immediate and late complications, and functional life. From January 2000 to August 2002, 187 PCs were placed in 187 patients (42% males, 51% Caucasians, mean age 50 +/- 14 years). There were no immediate complications related to the procedure such as hemorrhage, pulmonary embolism, or pneumothorax. There were a total of 35,078 catheter-days of follow-up. Sixteen catheters were removed during the observation period: three because of infection, seven after completion of chemotherapy, and six for other reasons. The remaining PCs are either functioning or the patients have died. The initial success rate was 100%. Kaplan-Meier analysis showed a 30-day survival of 97% and a 1-year survival of 92%. Interventional nephrologists, who have adequate training in central venous tunneled cuffed catheter placements, can successfully place PCs, with excellent success and minimal complications.

Successful orthotopic liver transplantation after trimethoprim-sulfamethoxazole associated fulminant liver failure.

Trimethoprim-sulfamethoxazole (TMP-SMZ) is one of the most commonly used antibiotics. Although many of its adverse effects are well recognized, TMP-SMZ related hepatotoxicity is considered rare and is usually characterized by cholestasis or mixed hepatocellular-holestatic reactions. In this study, we describe the case of a previously healthy young man with acute fulminant liver failure caused by TMP-SMZ. The patient presented with complaints of 'flu-like' symptoms with myalgia and fever after taking TMP-SMZ for 7 d for otitis externa. The patient subsequently developed fever, worsening jaundice, and a rash on his neck and chest. Liver enzymes peaked on day 3 with alanine aminotransferase (ALT) 11,549, aspartate aminotransferase (AST) 23,289, alkaline phosphatase 245, and total bilirubin 10.3 mg/dL, with a conjugated bilirubin of 8.3 mg/dL, prothrombin time (PT) 60.5 s, partial normalized ratio (PTT) 49 s, and international normalized ratio (INR) 7.5. Of note, acetaminophen level on admission was undetectable. Serology for hepatitis A, B, C, cytomegalovirus, HIV, toxoplasmosis, and blood cultures were all negative. The patient developed hepatic encephalopathy with hallucination on day 4. Laboratory tests revealed a serum ammonia level of 190 U, serum creatinine kinase (CK) 10,466 (42 on admission), serum creatinine 8.2 mg/dL (1.2 on admission), and significant metabolic acidosis. Renal ultrasound was unremarkable. The patient was started on hemodialysis for acute renal failure. Meanwhile, liver transplantation assessment was also initiated. On day 8 post-admission (15 d after taking TMP-SMZ), the patient received a successful orthotopic liver transplant.

Post transplant lymphocele: a single centre experience.

The occurrence of post renal transplant lymphocele is variable and the best approach to treatment is not well defined. The purpose of this study was to find out the incidence of post transplant lymphocele at our centre, identify demographic or surgical factors that may have influenced lymphocele formation, and distinguish the best approach to treatment. The charts of 138 consecutive renal transplant recipients from 1996 to 2001 were retrospectively reviewed. The demographic characteristics, comorbid illnesses, occurrence of lymphocele and its treatment modality were recorded. A total of 36 (26%) patients developed lymphoceles. There was a significant relationship between an increased body mass index (BMI) and lymphocele occurrence (P > 0.01). The recurrence rate with drainage alone was 33%, which decreased to 25% with sclerotherapy. In comparison, both laparoscopic and open surgical marsupialization had a much lower but similar recurrence rate of 12%. The laparoscopic method had less morbidity, a shortened hospital stay, and less infection than open surgery.