Automatic Detection of Oral Lesion Measurement Ruler Toward Computer-Aided Image-Based Oral Cancer Screening.

Intelligent computer-aided algorithms analyzing photographs of various mouth regions can help in reducing the high subjectivity in human assessment of oral lesions. Very often, in the images, a ruler is placed near a suspected lesion to indicate its location and as a physical size reference. In this paper, we compared two deep-learning networks: ResNeSt and ViT, to automatically identify ruler images. Even though the ImageN et 1K dataset contains a "ruler" class label, the pre-trained models showed low sensitivity. After fine-tuning with our data, the two networks achieved high performance on our test set as well as a hold-out test set from a different provider. Heatmaps generated using three saliency methods: GradCam and XRAI for ResNeSt model, and Attention Rollout for ViT model, demonstrate the effectiveness of our technique. Clinical Relevance- This is a pre-processing step in automated visual evaluation for oral cancer screening.

Oral Cavity Anatomical Site Image Classification and Analysis.

Oral cavity cancer is a common cancer that can result in breathing, swallowing, drinking, eating problems as well as speech impairment, and there is high mortality for the advanced stage. Its diagnosis is confirmed through histopathology. It is of critical importance to determine the need for biopsy and identify the correct location. Deep learning has demonstrated great promise/success in several image-based medical screening/diagnostic applications. However, automated visual evaluation of oral cavity lesions has received limited attention in the literature. Since the disease can occur in different parts of the oral cavity, a first step is to identify the images of different anatomical sites. We automatically generate labels for six sites which will help in lesion detection in a subsequent analytical module. We apply a recently proposed network called ResNeSt that incorporates channel-wise attention with multi-path representation and demonstrate high performance on the test set. The average F1-score for all classes and accuracy are both 0.96. Moreover, we provide a detailed discussion on class activation maps obtained from both correct and incorrect predictions to analyze algorithm behavior. The highlighted regions in the class activation maps generally correlate considerably well with the region of interest perceived and expected by expert human observers. The insights and knowledge gained from the analysis are helpful in not only algorithm improvement, but also aiding the development of the other key components in the process of computer assisted oral cancer screening.

The Development of Global Cancer Research at the United States National Cancer Institute.

International research and collaboration has been a part of the National Cancer Institute's (NCI) mission since its creation in 1937. Early on, efforts were limited to international exchange of information to ensure that US cancer patients could benefit from advances in other countries. As NCI's research grant portfolio grew in the 1950s, it included a modest number of grants to foreign institutions, primarily in the United Kingdom and Europe. In the 1960s, the development of geographic pathology, which aimed to study cancer etiology through variations in cancer incidence and risk factors, led to an increase in NCI-funded international research, including research in low- and middle-income countries. In this paper, we review key international research programs, focusing particularly on the first 50 years of NCI history. The first NCI-led overseas research programs, established in the 1960s in Ghana and Uganda, generated influential research but also struggled with logistical challenges and political instability. The 1971 National Cancer Act was followed by the creation of a number of bilateral agreements with foreign governments, including China, Japan, and Russia, to support cooperation in technology and medicine. Although these agreements were broad without specific scientific goals, they provided an important mechanism for sustained collaborations in specific areas. With the creation of the NCI Center for Global Health in 2011, NCI's global cancer research efforts gained sustained focus. Because the global cancer burden has evolved over time, increasingly impacting low- and middle-income countries, NCI's role in global cancer research remains more important than ever.

Perspectives on Strengthening Cancer Research and Control in Latin America Through Partnerships and Diplomacy: Experience of the National Cancer Institute's Center for Global Health.

According to the Pan American Health Organization, noncommunicable diseases, including cancer, are the leading causes of preventable and premature death in the Americas. Governments and health care systems in Latin America face numerous challenges as a result of increasing morbidity and mortality from cancer. Multiple international organizations have recognized the need for collaborative action on and technical support for cancer research and control in Latin America. The Center for Global Health at the US National Cancer Institute (NCI-CGH) is one entity among many that are working in the region and has sought to develop a strategy for working in Latin America that draws on and expands the collaborative potential of engaged, skilled, and diverse partners. NCI-CGH has worked toward developing and implementing initiatives in collaboration with global partners that share the common objectives of building a global cancer research community and translating research results into evidence-informed policy and practice. Both objectives are complementary and synergistic and are additionally supported by an overarching strategic framework that is focused on partnerships and science diplomacy. This work highlights the overall strategy for NCI-CGH engagement in Latin America through partnerships and diplomacy, and highlights selected collaborative efforts that are aimed at improving cancer outcomes in the region.

The Role of Affordable, Point-of-Care Technologies for Cancer Care in Low- and Middle-Income Countries: A Review and Commentary.

As the burden of non-communicable diseases such as cancer continues to rise in low- and middle-income countries (LMICs), it is essential to identify and invest in promising solutions for cancer control and treatment. Point-of-care technologies (POCTs) have played critical roles in curbing infectious disease epidemics in both high- and low-income settings, and their successes can serve as a model for transforming cancer care in LMICs, where access to traditional clinical resources is often limited. The versatility, cost-effectiveness, and simplicity of POCTs warrant attention for their potential to revolutionize cancer detection, diagnosis, and treatment. This paper reviews the landscape of affordable POCTs for cancer care in LMICs with a focus on imaging tools, in vitro diagnostics, and treatment technologies and aspires to encourage innovation and further investment in this space.

Defining a global research and policy agenda for betel quid and areca nut.

Betel quid and areca nut are known risk factors for many oral and oesophageal cancers, and their use is highly prevalent in the Asia-Pacific region. Additionally, betel quid and areca nut are associated with health effects on the cardiovascular, nervous, gastrointestinal, metabolic, respiratory, and reproductive systems. Unlike tobacco, for which the WHO Framework Convention on Tobacco Control provides evidence-based policies for reducing tobacco use, no global policy exists for the control of betel quid and areca nut use. Multidisciplinary research is needed to address this neglected global public health emergency and to mobilise efforts to control betel quid and areca nut use. In addition, future research is needed to advance our understanding of the basic biology, mechanisms, and epidemiology of betel quid and areca nut use, to advance possible prevention and cessation programmes for betel quid and areca nut users, and to design evidence-based screening and early diagnosis programmes to address the growing burden of cancers that are associated with use.

The National Institutes of Health Affordable Cancer Technologies Program: Improving Access to Resource-Appropriate Technologies for Cancer Detection, Diagnosis, Monitoring, and Treatment in Low- and Middle-Income Countries.

Point-of-care (POC) technologies have proved valuable in cancer detection, diagnosis, monitoring, and treatment in the developed world, and have shown promise in low-and-middle-income countries (LMIC) as well. Despite this promise, the unique design constraints presented in low-resource settings, coupled with the variety of country-specific regulatory and institutional dynamics, have made it difficult for investigators to translate successful POC cancer interventions to the LMIC markets. In response to this need, the National Cancer Institute has partnered with the National Institute of Biomedical Imaging and Bioengineering to create the National Institutes of Health Affordable Cancer Technologies (ACTs) program. This program seeks to simplify the pathway to market by funding multidisciplinary investigative teams to adapt and validate the existing technologies for cancer detection, diagnosis, and treatment in LMIC settings. The various projects under ACTs range from microfluidic cancer diagnostic tools to novel treatment devices, each geared for successful clinical adaptation to LMIC settings. Via progression through this program, each POC innovation will be uniquely leveraged for successful clinical translation to LMICs in a way not before seen in this arena.

Coregistration of preoperative computed tomography and intraoperative three-dimensional rotational x-ray images for cochlear implant surgical evaluation.

OBJECTIVE: A registration procedure of intraoperative three-dimensional rotational x-ray (3DRX) imaging and preoperative computed tomography (CT) imaging so that intraoperative CT quality imaging is available during cochlear implant surgery, providing detailed information concerning electrode position in the cochlea and its relation to surrounding bony structures. STUDY DESIGN: Retrospective case series SETTING: Tertiary referral center DATA: The imaging of five patients who had undergone cochlear implant surgery is used to develop a semiautomatic registration procedure to integrate intraoperative 3DRX and preoperative CT. The method is implemented in advanced medical imaging software to compute the transformations. The electrode is segmented from the registered 3DRX images using a semiautomated approach. The segmented electrode is superimposed onto the CT data. The methods are quantitatively validated based on expert-labeled anatomical landmarks. These landmarks are identified in the CT and 3DRX images by an expert. MAIN OUTCOME MEASURE: Mean error of the registration procedure for five anatomical landmarks in millimeters. RESULTS: Quantitative analysis showed a mean error of between 0.5 and 1 mm for all anatomical landmarks, suggesting that the results are trustworthy. CONCLUSION: We developed a reliable procedure for the registration of intraoperative 3DRX imaging and preoperative CT imaging for cochlear implant surgery. This registration procedure provides the ENT surgeon intraoperative high-quality CT imaging during cochlear implant surgery.

Mono- and multimodal registration of optical breast images.

Optical breast imaging offers the possibility of noninvasive, low cost, and high sensitivity imaging of breast cancers. Poor spatial resolution and a lack of anatomical landmarks in optical images of the breast make interpretation difficult and motivate registration and fusion of these data with subsequent optical images and other breast imaging modalities. Methods used for registration and fusion of optical breast images are reviewed. Imaging concerns relevant to the registration problem are first highlighted, followed by a focus on both monomodal and multimodal registration of optical breast imaging. Where relevant, methods pertaining to other imaging modalities or imaged anatomies are presented. The multimodal registration discussion concerns digital x-ray mammography, ultrasound, magnetic resonance imaging, and positron emission tomography.

Estimation of detection limits of a clinical fluorescence optical mammography system for the near-infrared fluorophore IRDye800CW: phantom experiments.

To evaluate if clinical fluorescence imaging of IRDye800CW is feasible on our fluorescence optical mammography system by estimating detection limits assessed by breast-cancer-simulating phantom experiments. Phantoms (2.1 cm(3), 0.9 cm(3)) with IRDye800CW concentrations of 0.5 to 120 nM were suspended in a 550 cm(3) measurement cup containing 507 surface-mounted source and detector fibers. The cup was filled with optical matching fluid containing IRDye800CW concentrations of 0, 5, 10, or 20 nM. Tomographic fluorescence images were acquired by exciting IRDye800CW at 730 nm; wavelengths above 750 nm were filtered. Signal intensities were calculated over a volume of interest corresponding to the size and location of the phantom in the reconstructed images. Correlations (R(2)) were calculated, and detection limits with associated upper 95% prediction interval were estimated. Between-day reproducibility was assessed with intraclass correlation coefficients (ICC). Fluorescent intensities were strongly correlated with phantom IRDye800CW concentrations (R(2)0.983 to 0.999). IRDye800CW detection limits ranged from 0.14 to 2.46 nM (upper 95% prediction limit 4.63 to 18.63 nM). ICC ranged from 0.88 to 1.00. The estimated detection limits for IRDye800CW were in the low-nanomolar range. These results support the start of clinical trials to evaluate the fluorescence optical mammography system using IRDye800CW labeled breast cancer targeting ligands.

Segmentation of 3D RF echocardiography using a multiframe spatio-temporal predictor.

We present an approach for segmenting left ventricular endocardial boundaries from RF ultrasound. Segmentation is achieved jointly using an independent identically distributed (i.i.d.) spatial model for RF intensity and a multiframe conditional model. The conditional model relates neighboring frames in the image sequence by means of a computationally efficient linear predictor that exploits spatio-temporal coherence in the data. Segmentation using the RF data overcomes problems due to image inhomogeneities often amplified in B-mode segmentation and provides geometric constraints for RF phase-based speckle tracking. The incorporation of multiple frames in the conditional model significantly increases the robustness and accuracy of the algorithm. Results are generated using between 2 and 5 frames of RF data for each segmentation and are validated by comparison with manual tracings and automated B-mode boundary detection using standard (Chan and Vese-based) level sets on echocardiographic images from 27 3D sequences acquired from 6 canine studies.

3D radio frequency ultrasound cardiac segmentation using a linear predictor.

We present an approach for segmenting the left ventricular endocardial boundaries from radio-frequency (RF) ultrasound. The method employs a computationally efficient two-frame linear predictor which exploits the spatio-temporal coherence of the data. By performing segmentation using the RF data we are able to overcome problems due to image inhomogeneities that are often amplified in B-mode segmentation, as well as provide geometric constraints for RF phase-based speckle tracking. We illustrate the advantages of our approach by comparing it to manual tracings of B-mode data and automated B-mode boundary detection using standard (Chan and Vese-based) level sets on echocardiographic images from 28 3D sequences acquired from 6 canine studies, imaged both at baseline and 1 hour post infarction.