Expert-Level Detection of Referable Glaucoma from Fundus Photographs in a Safety Net Population: The AI and Teleophthalmology in Los Angeles Initiative.

Purpose: To develop and test a deep learning (DL) algorithm for detecting referable glaucoma in the Los Angeles County (LAC) Department of Health Services (DHS) teleretinal screening program. Methods: Fundus photographs and patient-level labels of referable glaucoma (defined as cup-to-disc ratio [CDR] ≥ 0.6) provided by 21 trained optometrist graders were obtained from the LAC DHS teleretinal screening program. A DL algorithm based on the VGG-19 architecture was trained using patient-level labels generalized to images from both eyes. Area under the receiver operating curve (AUC), sensitivity, and specificity were calculated to assess algorithm performance using an independent test set that was also graded by 13 clinicians with one to 15 years of experience. Algorithm performance was tested using reference labels provided by either LAC DHS optometrists or an expert panel of 3 glaucoma specialists. Results: 12,098 images from 5,616 patients (2,086 referable glaucoma, 3,530 non-glaucoma) were used to train the DL algorithm. In this dataset, mean age was 56.8 ± 10.5 years with 54.8% females and 68.2% Latinos, 8.9% Blacks, 2.7% Caucasians, and 6.0% Asians. 1,000 images from 500 patients (250 referable glaucoma, 250 non-glaucoma) with similar demographics (p ≥ 0.57) were used to test the DL algorithm. Algorithm performance matched or exceeded that of all independent clinician graders in detecting patient-level referable glaucoma based on LAC DHS optometrist (AUC = 0.92) or expert panel (AUC = 0.93) reference labels. Clinician grader sensitivity (range: 0.33-0.99) and specificity (range: 0.68-0.98) ranged widely and did not correlate with years of experience (p ≥ 0.49). Algorithm performance (AUC = 0.93) also matched or exceeded the sensitivity (range: 0.78-1.00) and specificity (range: 0.32-0.87) of 6 LAC DHS optometrists in the subsets of the test dataset they graded based on expert panel reference labels. Conclusions: A DL algorithm for detecting referable glaucoma developed using patient-level data provided by trained LAC DHS optometrists approximates or exceeds performance by ophthalmologists and optometrists, who exhibit variable sensitivity and specificity unrelated to experience level. Implementation of this algorithm in screening workflows could help reallocate eye care resources and provide more reproducible and timely glaucoma care.

Urine Multi-drug Screening with GC-MS or LC-MS-MS Using SALLE-hybrid PPT/SPE.

To intoxicated patients in the emergency room, toxicological analysis can be considerably helpful for identifying the involved toxicants. In order to develop a urine multi-drug screening (UmDS) method, gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-tandem mass spectrometry (LC-MS-MS) were used to determine targeted and unknown toxicants in urine. A GC-MS method in scan mode was validated for selectivity, limit of detection (LOD) and recovery. An LC-MS-MS multiple reaction monitoring (MRM) method was validated for lower LOD, recovery and matrix effect. The results of the screening analysis were compared with patient medical records to check the reliability of the screen. Urine samples collected from an emergency room were extracted through a combination of salting-out assisted liquid-liquid extraction (SALLE) and hybrid protein precipitation/solid phase extraction (hybrid PPT/SPE) plates and examined by GC-MS and LC-MS-MS. GC-MS analysis was performed as unknown drug screen and LC-MS-MS analysis was conducted as targeted drug screen. After analysis by GC-MS, a library search was conducted using an in-house library established with the automated mass spectral deconvolution and identification system (AMDISTM). LC-MS-MS used Cliquid®2.0 software for data processing and acquisition in MRM mode. An UmDS method by GC-MS and LC-MS-MS was developed by using a SALLE-hybrid PPT/SPE and in-house library. The results of UmDS by GC-MS and LC-MS-MS showed that toxicants could be identified from 185 emergency room patient samples containing unknown toxicants. Zolpidem, acetaminophen and citalopram were the most frequently encountered drugs in emergency room patients. The UmDS analysis developed in this study can be used effectively to detect toxic substances in a short time. Hence, it could be utilized in clinical and forensic toxicology practices.

Prevalence of new psychoactive substances in Northeast Asia from 2007 to 2015.

The proliferation of new psychoactive substances (NPS) has been a global trend in drug abuse and its regulation has been a worldwide concern. There is no doubt that it is necessary to share information related to these emerging substances between countries and continents for the effective regulation of NPS. With efforts for the efficient regulation of NPS, many studies and information have been published for the prevalence of NPS in the United States and other countries in Europe and Oceania. However, there is lack of information available for the prevalence of NPS in Asian and African countries. Therefore, this research was focused on the investigation of legal status of certain NPS in Northeast Asian countries, including China, Japan, South Korea and Taiwan, in order to provide information on the prevalence and trend of emerging NPS in these countries. The results showed that a total of 940 NPS was reported in 4 Northeast Asian countries from 2007 to 2015. Among 940 NPS, 882 NPS are legally restricted in at least one country (94%) and 96 substances were not currently under control (6%) in these countries. The number of controlled NPS that are currently controlled in all 4 countries was only 25 (or 28%) out of 882 NPS. Each substance was categorized in 9 groups according to the classification proposed by the United Nations Office on Drugs and Crime (UNODC). In Northeast Asia, the most commonly controlled NPS were synthetic cannabinoids, synthetic cathinones, and phenethylamines. It was found that Japan is the most proactive country in terms of the NPS regulation with 41% of the total number of controlled NPS in Northeast Asia, followed by South Korea (21%), China (28%), Taiwan (10%). Comparing the number of NPS newly regulated in each country every year, NPS has been broadly scheduled in 2011 and the number of scheduled NPS has dramatically increased from 2013 to 2015. It was shown that Northeast Asia is also in danger of these emerging NPS and the effective regulation across countries is important for the prevention of NPS. Also, this study will bring attention to local law enforcement in the construction of local drug crime prevention network sharing information for these controlled substances.