Ilwoo Park

Bio: Ilwoo Park is an academic researcher from Chonnam National University. The author has contributed to research in topics: Magnetic resonance spectroscopic imaging & Magnetic resonance imaging. The author has an hindex of 21, co-authored 43 publications receiving 2683 citations. Previous affiliations of Ilwoo Park include University of California, San Francisco.

Metabolic Imaging of Patients with Prostate Cancer Using Hyperpolarized [1-13C]Pyruvate

Abstract: This first-in-man imaging study evaluated the safety and feasibility of hyperpolarized [1-13C]pyruvate as an agent for noninvasively characterizing alterations in tumor metabolism for patients with prostate cancer. Imaging living systems with hyperpolarized agents can result in more than 10,000-fold enhancement in signal relative to conventional magnetic resonance (MR) imaging. When combined with the rapid acquisition of in vivo 13C MR data, it is possible to evaluate the distribution of agents such as [1-13C]pyruvate and its metabolic products lactate, alanine, and bicarbonate in a matter of seconds. Preclinical studies in cancer models have detected elevated levels of hyperpolarized [1-13C]lactate in tumor, with the ratio of [1-13C]lactate/[1-13C]pyruvate being increased in high-grade tumors and decreased after successful treatment. Translation of this technology into humans was achieved by modifying the instrument that generates the hyperpolarized agent, constructing specialized radio frequency coils to detect 13C nuclei, and developing new pulse sequences to efficiently capture the signal. The study population comprised patients with biopsy-proven prostate cancer, with 31 subjects being injected with hyperpolarized [1-13C]pyruvate. The median time to deliver the agent was 66 s, and uptake was observed about 20 s after injection. No dose-limiting toxicities were observed, and the highest dose (0.43 ml/kg of 230 mM agent) gave the best signal-to-noise ratio for hyperpolarized [1-13C]pyruvate. The results were extremely promising in not only confirming the safety of the agent but also showing elevated [1-13C]lactate/[1-13C]pyruvate in regions of biopsy-proven cancer. These findings will be valuable for noninvasive cancer diagnosis and treatment monitoring in future clinical trials.

Hyperpolarized 13C magnetic resonance metabolic imaging: application to brain tumors

Abstract: In order to compare in vivo metabolism between malignant gliomas and normal brain, 13C magnetic resonance (MR) spectroscopic imaging data were acquired from rats with human glioblastoma xenografts (U-251 MG and U-87 MG) and normal rats, following injection of hyperpolarized [1-13C]-pyruvate. The median signal-to-noise ratio (SNR) of lactate, pyruvate, and total observed carbon-13 resonances, as well as their relative ratios, were calculated from voxels containing Gadolinium-enhanced tissue in T1 postcontrast images for rats with tumors and from normal brain tissue for control rats. [1-13C]-labeled pyruvate and its metabolic product, [1-13C]-lactate, demonstrated significantly higher SNR in the tumor compared with normal brain tissue. Statistical tests showed significant differences in all parameters (P < .0004) between the malignant glioma tissue and normal brain. The SNR of lactate, pyruvate, and total carbon was observed to be different between the U-251 MG and U-87 MG models, which is consistent with inherent differences in the molecular characteristics of these tumors. These results suggest that hyperpolarized MR metabolic imaging may be valuable for assessing prognosis and monitoring response to therapy for patients with brain tumors.

Deep neural networks show an equivalent and often superior performance to dermatologists in onychomycosis diagnosis: Automatic construction of onychomycosis datasets by region-based convolutional deep neural network.

Abstract: Although there have been reports of the successful diagnosis of skin disorders using deep learning, unrealistically large clinical image datasets are required for artificial intelligence (AI) training. We created datasets of standardized nail images using a region-based convolutional neural network (R-CNN) trained to distinguish the nail from the background. We used R-CNN to generate training datasets of 49,567 images, which we then used to fine-tune the ResNet-152 and VGG-19 models. The validation datasets comprised 100 and 194 images from Inje University (B1 and B2 datasets, respectively), 125 images from Hallym University (C dataset), and 939 images from Seoul National University (D dataset). The AI (ensemble model; ResNet-152 + VGG-19 + feedforward neural networks) results showed test sensitivity/specificity/ area under the curve values of (96.0 / 94.7 / 0.98), (82.7 / 96.7 / 0.95), (92.3 / 79.3 / 0.93), (87.7 / 69.3 / 0.82) for the B1, B2, C, and D datasets. With a combination of the B1 and C datasets, the AI Youden index was significantly (p = 0.01) higher than that of 42 dermatologists doing the same assessment manually. For B1+C and B2+ D dataset combinations, almost none of the dermatologists performed as well as the AI. By training with a dataset comprising 49,567 images, we achieved a diagnostic accuracy for onychomycosis using deep learning that was superior to that of most of the dermatologists who participated in this study.

High-performance medicine: the convergence of human and artificial intelligence

Abstract: The use of artificial intelligence, and the deep-learning subtype in particular, has been enabled by the use of labeled big data, along with markedly enhanced computing power and cloud storage, across all sectors. In medicine, this is beginning to have an impact at three levels: for clinicians, predominantly via rapid, accurate image interpretation; for health systems, by improving workflow and the potential for reducing medical errors; and for patients, by enabling them to process their own data to promote health. The current limitations, including bias, privacy and security, and lack of transparency, along with the future directions of these applications will be discussed in this article. Over time, marked improvements in accuracy, productivity, and workflow will likely be actualized, but whether that will be used to improve the patient-doctor relationship or facilitate its erosion remains to be seen.

The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions

Abstract: Training of neural networks for automated diagnosis of pigmented skin lesions is hampered by the small size and lack of diversity of available datasets of dermatoscopic images. We tackle this problem by releasing the HAM10000 ("Human Against Machine with 10000 training images") dataset. We collected dermatoscopic images from different populations acquired and stored by different modalities. Given this diversity we had to apply different acquisition and cleaning methods and developed semi-automatic workflows utilizing specifically trained neural networks. The final dataset consists of 10015 dermatoscopic images which are released as a training set for academic machine learning purposes and are publicly available through the ISIC archive. This benchmark dataset can be used for machine learning and for comparisons with human experts. Cases include a representative collection of all important diagnostic categories in the realm of pigmented lesions. More than 50% of lesions have been confirmed by pathology, while the ground truth for the rest of the cases was either follow-up, expert consensus, or confirmation by in-vivo confocal microscopy.

Metabolic Imaging of Patients with Prostate Cancer Using Hyperpolarized [1-13C]Pyruvate

Abstract: This first-in-man imaging study evaluated the safety and feasibility of hyperpolarized [1-13C]pyruvate as an agent for noninvasively characterizing alterations in tumor metabolism for patients with prostate cancer. Imaging living systems with hyperpolarized agents can result in more than 10,000-fold enhancement in signal relative to conventional magnetic resonance (MR) imaging. When combined with the rapid acquisition of in vivo 13C MR data, it is possible to evaluate the distribution of agents such as [1-13C]pyruvate and its metabolic products lactate, alanine, and bicarbonate in a matter of seconds. Preclinical studies in cancer models have detected elevated levels of hyperpolarized [1-13C]lactate in tumor, with the ratio of [1-13C]lactate/[1-13C]pyruvate being increased in high-grade tumors and decreased after successful treatment. Translation of this technology into humans was achieved by modifying the instrument that generates the hyperpolarized agent, constructing specialized radio frequency coils to detect 13C nuclei, and developing new pulse sequences to efficiently capture the signal. The study population comprised patients with biopsy-proven prostate cancer, with 31 subjects being injected with hyperpolarized [1-13C]pyruvate. The median time to deliver the agent was 66 s, and uptake was observed about 20 s after injection. No dose-limiting toxicities were observed, and the highest dose (0.43 ml/kg of 230 mM agent) gave the best signal-to-noise ratio for hyperpolarized [1-13C]pyruvate. The results were extremely promising in not only confirming the safety of the agent but also showing elevated [1-13C]lactate/[1-13C]pyruvate in regions of biopsy-proven cancer. These findings will be valuable for noninvasive cancer diagnosis and treatment monitoring in future clinical trials.