Machine Learning Methods to Investigate Drug Delivery in Infusion Pump

In the current scenario, usage of the smart medical pump is predominant in the medical field. The precise drug dosage, flow accuracy should be maintained to increase the performance of an infusion pump. In this work, an attempt has been made to predict and control the speed of the infusion pump for suitable infusion flowrate using machine learning technique and Linear Quadratic Gaussian (LQG) controller. The data for this study is considered from the publicly available online database, electronic Medicines Compendium (eMC). The speed of the infusion pump has been calculated using the drug dosage and flow rate for two different drugs. The prediction of infusion pump speed is achieved using Linear regression with Principal Component analysis (PCR) and Support Vector Machine Regression (SVR). The performance of the prediction schemes is evaluated using standard metrics. To validate the optimal control of the predicted speed, two different medical graded motors are considered. Further, the optimal control of the pump speed is investigated using Proportional–Integral–Derivative (PID), Linear Quadratic Regulator (LQR), and LQG controllers for its stability criteria. The prediction of the pump speed using regression models PCR, SVR has been verified and then the transient response analysis with rise time, settling time for both the motors have been examined. Results demonstrate that the LQG optimal control strategy achieves fast rise time, settling time of motor1 with 0.653s, 1.15s, and 0.22, 0.392s for motor2 respectively.

Self-Perceived Skills of Pre Intern Doctors in Sri Lanka

IntroductionThe transition from medical student to intern is a significant and challenging time in a doctor’s training. It can be stressful for the young doctor and there can be varying expectations from employers and supervisors on their capabilities. There is a time gap between graduation and internship for medical graduates in Sri Lanka. Also, there was no proper orientation prior the internship. Thus, the Good Intern Programme was developed as a skills training programme for pre interns who are awaiting internship, and this was conducted in collaboration with Ministry of Health (MOH), Government Medical Officers’ Association (GMOA), Society for Health Research and Innovation (SHRI). This study aimed to explore self-perceived competency of selected skills of pre-interns who were awaiting for internship. Methods Study populationSri Lankan medical Faculties (University of Colombo, Peradeniya, Sri Jayewardenapura, Jaffna, Kelaniya, Ruhuna, Eastern University- Batticaloa, Rajarata University and General Sir John Kotelawala Defence University -KDU), and foreign universities produce approximately 1000 medical graduates per year. In 2020, all pre- interns who joined the Good Intern Programme, Sri Lanka were invited to participate in this questionnaire survey. Statistical analyses were performed using SPSS 23.0 software. Descriptive data were presented as proportions.Results Mean age was 27.6 (+ SD 1.4). Majority of the pre-interns were females (66.5%). Majority of pre-interns perceived that they were able to performs suturing, cannulation, catheterization, setting up an Intravenous (IV) drip, infusion pump setup, blood and blood products transfusion, venipuncture, venipuncture for blood culture, injections-Subcutaneous (SC)/Intramuscular (IM), Glasgow Coma Scale (GCS) monitoring, Cardiopulmonary resuscitation (CPR), arterial puncture for Arterial Blood Gases (ABG), wound dressing, suture removal, bandaging, glucometer use, nebulization, connecting to an ECG (electrocardiogram) monitor competently with or without supervision respectively. Lack of competency perceived on the following skills. Nasogastric tube (NG) insertion, pleural tap, peritoneal tap, removal of an Intercostal tube, lumbar puncture, defibrillation, venous cut down, intubation, CVP (central venous catheter) line insertion, Intercostal (IC) tube insertion were rated by the majority of pre-interns as skills that could be performed competently with supervision or not able to perform the skill but has observed skill. ConclusionAlthough most of the skills were rated by majority of the pre-interns as skills that could be competently performed with or without supervision respectively, there were some skills which needed some improvement. This study suggests that Good Intern Programme in pre-intern period can help to prepare students for the intern role.

Fatal accidental lipid overdose with intravenous composite lipid emulsion in a premature newborn: a case report

Background Tenfold or more overdose of a drug or preparation is a dreadful adverse event in neonatology, often due to an error in programming the infusion pump flow rate. Lipid overdose is exceptional in this context and has never been reported during the administration of a composite intravenous lipid emulsion (ILE). Case presentation Twenty-four hours after birth, a 30 weeks’ gestation infant with a birthweight of 930 g inadvertently received 28 ml of a composite ILE over 4 h. The ILE contained 50% medium-chain triglycerides and 50% soybean oil, corresponding to 6 g/kg of lipids (25 mg/kg/min). The patient developed acute respiratory distress with echocardiographic markers of pulmonary hypertension and was treated with inhaled nitric oxide and high-frequency oscillatory ventilation. Serum triglyceride level peaked at 51.4 g/L, 17 h after the lipid overload. Triple-volume exchange transfusion was performed twice, decreasing the triglyceride concentration to < 10 g/L. The infant’s condition remained critical, with persistent bleeding and shock despite supportive treatment and peritoneal dialysis. Death occurred 69 h after the overdose in a context of refractory lactic acidosis. Conclusions Massive ILE overdose is life-threatening in the early neonatal period, particularly in premature and hypotrophic infants. This case highlights the vigilance required when ILEs are administered separately from other parenteral intakes. Exchange transfusion should be considered at the first signs of clinical or biological worsening to avoid progression to multiple organ failure.

298 Domiciliary periodic Levosimendan infusions in advanced heart failure: a single centre experience

Aims There are several studies that describes a possible use of periodical infusions of Levosimendan in patients with advance heart failure (AHF). In these studies, Levosimendan was administered in ambulatory settings (LevoRep, LION-HEART) or during 24 h of hospital stay (LAICA) and for a maximum of 12 months. Currently there is no strong evidence about the feasibility and safety of performing the infusions in an out of hospital setting. Aim of our study was to analyse the feasibility and safety of periodical Levosimendan infusions in AHF patients, in an out of hospital setting for palliative use. Methods and results Aretrospective study was conducted from June 2016 to September 2021 on AHF patients, followed by our centre for periodic infusion of Levosimendan. All patients were previously implanted with an ICD, were ineligible for heart transplant or for ventricular assist devices. All patients on each cycle underwent a cardiological evaluation, a blood sample (complete blood count, biochemical profile, coagulation profile, and NTproBNP), and Levosimendan was administrated with a wearable infusion pump (CADD-Legacy®; Smiths Medical). The patient went home and came back the day after, to return the pump and for reassessment. We collected data on the different infusion regimen, adapted for every patient, and on adverse events during infusions. We enrolled 11 patients with AHF, 9 (82%) of which of ischaemic origin, 2 (18%) were female, median age 72 years (IQR: 64–75), 10 with NYHA class III, 1 with a NYHA class IV, for a total of 231 infusions, with a median Follow-up of 459 days (IQR 132; 783). A total dose of 6.25 mg was administered at 0.05 μg/kg/min in 24 h. In five patients the infusion was of 12.5 mg at 0.1 μg/kg/min rate in 24 h and in other three patients at 0.05 μg/kg/min in 48 h. The median interval between administrations was 21 days (IQR: 14–24.5). Mild adverse events (i.e. that did not contraindicate successive cycles) were 7 on 231 infusions (3.0%): symptomatic hypotension (four cases, 1.7%), loss of consciousness (two cases 0.9%), bleeding for vein access misplacement (one case, 0.4%). No serious adverse event such as sustained or malignant arrhythmia, allergic reaction, or need for hospital admission were recorded. Conclusions The administration of Levosimendan at home with a wearable iv pump is safe and well tolerated. This practice allows to reduce infusion costs of day hospital regime and to lengthen the time available for drug infusion, thus reducing the events caused by a too rapid infusion.