Clinical evaluation of a deep-learning-based computer-aided detection system for the detection of pulmonary nodules in a large teaching hospital.

AIM: To evaluate a deep-learning-based computer-aided detection (DL-CAD) software system for pulmonary nodule detection on computed tomography (CT) images and assess its added value in the clinical practice of a large teaching hospital. MATERIALS AND METHODS: A retrospective analysis was performed of 145 chest CT examinations by comparing the output of the DL-CAD software with a reference standard based on the consensus reading of three radiologists. For every nodule in each scan, the location, composition, and maximum diameter in the axial plane were recorded. The subgroup of chest CT examinations (n = 97) without any nodules was used to determine the negative predictive value at the given clinical sensitivity threshold setting. RESULTS: The radiologists found 91 nodules and the CAD system 130 nodules of which 80 were true positive. The measured sensitivity was 88% and the mean false-positive rate was 1.04 false positives/scan. The negative predictive value was 95%. For 23 nodules, there was a size discrepancy of which 19 (83%) were measured smaller by the radiologist. The agreement of nodule composition between the CAD results and the reference standard was 95%. CONCLUSIONS: The present study found a sensitivity of 88% and a false-positive rate of 1.04 false positives/scan, which match the vendor specification. Together with the measured negative predictive value of 95% the system performs very well; however, these rates are still not good enough to replace the radiologist, even for the specific task of nodule detection. Furthermore, a surprisingly high rate of overestimation of nodule size was observed, which can lead to too many follow-up examinations.

Complications and failure of uterine artery embolisation for intractable postpartum haemorrhage.

OBJECTIVE: Primary postpartum haemorrhage (PPH) is a major cause of maternal morbidity and mortality around the world. Most patients can be managed conservatively, but patients with intractable bleeding require more aggressive treatment. In these cases uterine artery embolisation (UAE) has proven to be a useful tool to control PPH. The reported success rate of UAE is over 90% with only minor complications. In this case series we studied the effectiveness and complications of UAE. DESIGN: Retrospective analysis of a case series. SETTING: Case series in a large peripheral hospital in the Netherlands. SAMPLE: Eleven patients who were treated with UAE for intractable PPH from November 2004 to February 2008. METHODS: In this paper we review the results of all patients treated with UAE for intractable PPH in our hospital and focus on the two cases with adverse outcomes. MAIN OUTCOME MEASURES: Effectiveness, causes of failure of UAE, complications. RESULTS: Nine out of eleven patients were treated successfully with UAE. One patient needed an emergency hysterectomy for intractable bleeding. In the aftermath she developed a vesicovaginal fistula (VVF). Another patient suffered a major thrombo-embolic event of the right leg, for which she underwent embolectomies and despite fasciotomy a necrotectomy. CONCLUSIONS: UAE is a valuable tool in managing major PPH and in most cases it can replace surgery and thus prevent sacrification of the uterus. However, due to blood supply of the uterus by one of the ovarian or aberrant arteries, UAE might fail to control the bleeding. In addition, serious complications such as a thrombo-embolic event or VVF may occur. We hereby present a case of migration of an embolus from the site of re-embolisation into the femoral artery requiring immediate intervention to prevent the loss of the lower leg. This complication demonstrates that gelatine sponge particles could migrate from the internal iliac artery into the external iliac artery.

Safety and long-term effects of renal denervation: Rationale and design of the Dutch registry.

BACKGROUND: Percutaneous renal denervation (RDN) has recently been introduced as a treatment for therapy-resistant hypertension. Also, it has been suggested that RDN may be beneficial for other conditions characterised by increased sympathetic nerve activity. There are still many uncertainties with regard to efficacy, safety, predictors for success and long-term effects. To answer these important questions, we initiated a Dutch RDN registry aiming to collect data from all RDN procedures performed in the Netherlands. METHODS: The Dutch RDN registry is an ongoing investigator-initiated, prospective, multicentre cohort study. Twenty-six Dutch hospitals agreed to participate in this registry. All patients who undergo RDN, regardless of the clinical indication or device that is used, will be included. Data are currently being collected on eligibility and screening, treatment and follow-up. RESULTS: Procedures have been performed since August 2010. At present, data from 306 patients have been entered into the database. The main indication for RDN was hypertension (n = 302, 99%). Patients had a mean office blood pressure of 177/100 (±29/16) mmHg with a median use of three (range 0-8) blood pressure lowering drugs. Mean 24-hour blood pressure before RDN was 157/93 (±18/13) mmHg. RDN was performed with different devices, with the Simplicity™ catheter currently used most frequently. CONCLUSION: Here we report on the rationale and design of the Dutch RDN registry. Enrolment in this investigator-initiated study is ongoing. We present baseline characteristics of the first 306 participants.