ABSTRACT
The inverse problem of diffuse optical tomography (DOT) is ill-posed. Traditional method cannot achieve high imaging accuracy and the calculation process is time-consuming, which restricts the clinical application of DOT. Therefore, a method based on stacked auto-encoder (SAE) was proposed and used for the DOT inverse problem. Firstly, a traditional SAE method is used to solved the inverse problem. Then, the output structure of SAE neural network is improved to a single output SAE, which reduce the burden on the neural network. Finally, the improved SAE method is used to compare with traditional SAE method and traditional levenberg-marquardt (LM) iterative method. The result shows that the average time to solve the inverse problem of the method proposed in this paper is only 1.67% of the LM method. The mean square error (MSE) value is 46.21% lower than the traditional iterative method, 61.53% lower than the traditional SAE method, and the image correlation coefficient(ICC) value is 4.03% higher than the traditional iterative method, 18.7% higher than the traditional SAE method and has good noise immunity under 3% noise conditions. The research results in this article prove that the improved SAE method has higher image quality and noise resistance than the traditional SAE method, and at the same time has a faster calculation speed than the traditional iterative method, which is conducive to the application of neural networks in DOT inverse problem calculation.
Subject(s)
Algorithms , Neural Networks, Computer , Tomography, OpticalABSTRACT
Fluorescent Diffuse Optical Tomography (FDOT) is an emerging imaging method with great prospects in fields of biology and medicine. However, the current solutions to the forward problem in FDOT are time consuming, which greatly limit the application. We proposed a method for FDOT based on Lattice Boltzmann forward model on GPU to greatly improve the computational efficiency. The Lattice Boltzmann Method (LBM) was used to construct the optical transmission model. This method separated the LBM into collision, streaming and boundary processing processes on GPUs to perform the LBM efficiently, which were local computational and inefficient on CPU. The feasibility of the proposed method was verified by the numerical phantom and the physical phantom experiments. The experimental results showed that the proposed method achieved the best performance of a 118-fold speed up under the precondition of simulation accuracy, comparing to the diffusion equation implemented by Finite Element Method (FEM) on CPU. Thus, the LBM on the GPU may efficiently solve the forward problem in FDOT.
Subject(s)
Computers , Fluorescence , Phantoms, Imaging , Tomography, Optical/methodsABSTRACT
Fluorescence Diffuse Optical Tomography (FDOT) is significant for biomedical applications, such as medical diagnostics, drug research. The fluorescence probe distribution in biological tissues can be quantitatively and non-invasively obtained via FDOT, achieving targets positioning and detection. In order to reduce the cost of FDOT, this study designs a FDOT system based on Lattice Boltzmann forward model. The system is used to realize two functions of light propagation simulation and FDOT reconstruction, and is composed of a parameter module, an algorithm module, a result display module and a data interaction module. In order to verify the effectiveness of the platform, this study carries out the light propagation simulation experiment and the FDOT reconstruction experiment, respectively comparing the Monte Carlo (MC) light propagation simulation results and the real position of the light source to be reconstructed. Experiments show that the proposed FDOT system has good reliability and has a high promotion value.
Subject(s)
Algorithms , Computer Simulation , Monte Carlo Method , Optical Devices , Reproducibility of Results , Tomography, OpticalABSTRACT
Objective To investigate the clinical value of conventional ultrasonographic ( US),US-guided diffuse optical tomography ( US-guided DOT ) and both combined to assess treatment response of breast cancer to neoadjuvant chemotherapy( NAC). Methods Eighty-eight breast cancer patients,totally 93 lesions were included in the study. Pre-and post-last chemotherapy,size,and total hemoglobin concentration ( THC) of each lesion were measured by conventional US and US-guided DOT before biopsy,the change of lesion Size( ΔSize) and the change of THC( ΔTHC) were calculated respectively.Based on the guidelines to evaluate the response to treatment in solid tumors,the responses to NAC were classified into complete response, partial response, static disease, progressive disease groups. The histological response to chemotherapy were categorised as partial pathological response and complete pathological response using Miller and Payne system. Results Of 93 breast cancers,the overall response rate was 81.7%,the cPR rate was 24.7%. According to ROC curve analysis,when ΔSize 42.6% and ΔTHC 23.9% as cutoff values to evaluate the complete response and partial response,the area under the curve ( AUC ) were 0.666 and 0.751,respectively,the AUC of US and US-guided DOT combined in parallel and in series were 0.680 and 0.737 respectively.When ΔSize 64.5% and ΔTHC 27.2% as cutoff values to evaluate complete pathologial response,the AUC were 0.690 and 0.728 respectively,the AUC of US and US-guided DOT combined in parallel and in series were 0.693 and 0.726 respectively. Conclusions US-guided DOT and US can be used to predict breast cancers response to NAC,US and US-guided DOT combined in parallel and in series can not improve response prediction comparing with US or US-guided DOT alone.
ABSTRACT
Objective To discuss the value of early prediction and evaluation using ultrasound-guided diffuse optical tomography (US-guided DOT) of neoadjuvant chemotherapy (NCT) for breast cancer.Methods Sixty-two primary breast cancer patients were included.Before every neoadjuvant chemotherapy and surgery,the total hemoglobin concentration(Hb) was measured by US-guided DOT.According to the RECIST 1.1 standard,62 patients were divided into partial response (PR) group,complete response (CR) group,progressive disease(PD) group and stable disease(SD) group.According to the Miller and Payne (MP) grading system,the patients were divided into poor efficacy group(MP grade 1-3) and the effect of excellent group (MP grade 4-5).According to the postoperative pathological results,patients were divided into pathologic complete response(pCR) group and non pCR group.Hb with imaging and postoperative pathological results were analyzed.Results All of the 62 patients finished the NCT.After first cycle of chemotherapy and all chemotherapy,the value of Hb in the above groups had different degrees of decline,and the difference among the groups was statistically significant(P ＜0.001).When the descending value of Hb was 44.5 as the threshold for early prediction of NCT efficiency,the area under the curve of ROC (Az) was 0.745,and the sensitivity was 75.0％,specificity was 73.1％,positive predictive value was 79.4％,negative predictive value was 67.9％,and accuracy was 74.2％,respectively.Conclusions In the process of neoadjuvant chemotherapy of breast cancer,the decrease of Hb value is positively related to the therapeutic effect of chemotherapy.The change of Hb value can predict and evaluate the efficacy of neoadjuvant chemotherapy in the early time.
ABSTRACT
Objective To explore the ultrasound-guided diffuse optical tomography (DOT) distribution difference and clinical characteristics of T 1-T3 breast carcinoma. Methods Four hundred and forty-seven breast cancer patients with 455 breast lesions were enrolled.The lesion maximal diameter and total hemoglobin concentration(THC) were obtained by ultrasound-guided DOT before breast surgery biopsy.The patients age,body mass index(BMI),and the distances from lesion to skin and nipple were measured,lymph node status were also assessed.According to the 7th Edition of the AJCC Cancer Staging, all tumors were divided into T1( ≤ 2 cm),T2(2 cm < lesion ≤ 5 cm),and T3( > 5 cm) three groups according to maximum diameter on ultrasound.Results Out of 455 breast cancer lesions,148 lesions were stage T1,251 lesions were stage T2,56 lesions were T3.The age,lesion size,lesion THC and the distances between lesions and nipples of breast cancer patients were found that there were significant differences among T1,T2 and T3(all P = 0.000). With T stages increasing,the age of breast cancers patients decreased,the distance between lesions and nipples decreased,the THC increased.THC of breast cancer T 2 [(221.0 ± 56.0)μmol/L] and T3[(233.1 ± 54.0)μmol/L] were significant higher than that of T1 [(181.6 ± 70.4)μmol/L](all P = 0.000).There were all no significant difference for BMI and the distancefrom lesion to skin among different stages T( P > 0.05).With T stages increasing,the ratio of lymph node metastasis increased(T1 26.4%,T2 49.8%,T3 55.4%; P =0.000).Conclusions With breast cancer T stages increasing,T HC increases,the age of the patients decreases,lymph node metastasis rate increases.
ABSTRACT
Objective To investigate the value of ultrasound-guided diffuse optical tomography (US-guided DOT) to predict clinical efficacy of breast cancer neoadjuvant chemotherapy (NAC).Methods Eighty-eight breast cancer patients with 93 lesions were included.Pre-and post-last chemotherapy,the size,total hemoglobin concentration (THC) of each lesion were measured by ultrasonography (US) and US-guided DOT.Based on the guidelines to evaluate the response to treatment in solid tumors,the lesions of treated breast cancer patients were divided into 4 types of responses to NAC:complete response (CR),partial response (PR),stable disease (SD),and progressive disease (PD).Efficient groups include CR and PR groups.Results As expected,no significant difference was found in size and THC for untreated lesions (all P ＞0.05).However,for the treated lesions(P =0.001),THC,pre-vs post-treatment size changes (△Size％) (P =0.002) and THC changes (△THC％) (P ＜0.001) were significantly varied among CR,PR,SD,PD groups.When compared with pre-treated,tumor sizes after treatment were changed significantly in all CR(P ＜0.001),PR(P ＜0.001),SD (P =0.023) and PD (P =0.001),while significant change of THC was only found in CR(P ＜0.001),PR(P ＜0.001) and SD (P =0.002).When △THC％ =23.9％ as the threshold for prediction of NAC efficiency,the area under the curve of ROC was 0.75,and the sensitivity was 73.7％,specificity was 76.5 ％,positive predictive value was 93.3 ％,negative predictive value was 39.4％,accuracy was 74.2％.Conclusions △size％ changes in consistent with △THC％ among the intergroups,but their changes levels are different,the highest change percent appears in CR,gradually decreased in PR,SD groups.△THC％ will contribute to predict preoperative clinical NAC efficacy.
ABSTRACT
Objective To investigate the value of ultrasound-diffuse optical tomography(US-DOT) in evaluation of the response to neoadjuvant chemotherapy (NCT) in patients with breast cancer.Methods Fifty women with breast cancer confirmed by biopsy-pathology scheduled to undergo NCT were enrolled.US-DOT was performed and the total hemoglobin concentration(HbT) prior to and at the end of NCT.According to the response to treatment in solid tumors,patients were divided into complete response(CR) group,partial response (PR) group,stable disease (SD) group and progressive disease (PD) group.The analysis of variance compared the differences of HbT before and after NCT in each group; Pearson correlation analyzed the relationship between the relative HbT variation and the tumor size changes;Mixedeffects model analyzed the relationship between HbT and tumor size.Results After NCT,CR group had 8 cases,PR group 30 cases,SD group 12 cases and no PD cases.Before and after NCT,the mean relative variations of HbT before and after NCT were-0.525 ± 0.222,-0.328 ± 0.180 and-0.173 ± 0.167 in CR,PR and SD groups,respectively.The differences of HbT before and after NCT were statistically significant in each group (P ＜0.05).There was significant difference among tumor diameter and HbT in CR group and PR group (P ＜0.001);while in the SD group,there was no difference among tumor diameter and HbT (P ＞ 0.05).The change of HbT after NCT showed positive correlation with the change in tumor size (r =0.6).Mixed-effects model also showed that HbT change was associated with tumor size.Conclusions USDOT marker HbT is closely related to tumor size of breast cancer before and after NCT and may be applied to evaluate the response of breast cancer to NCT.
ABSTRACT
Based on photon counting mode,a multi-channel continuous-wave(CW) diffuse optical tomography(DOT) system is designed for galactophore inspection.For the acquisition of dynamic information,the instrument adopts the non-intermittent counters to achieve the function of width-adjustable moving gate.The width of the moving gate can be changed from 50 microseconds to about 10 seconds,and correspondingly the total temporal span of the measurement ranges from several milliseconds to tens seconds,depending on the preset gate number.The data transfer as well as the system control is through the universal serial bus(USB) 2.0 technique,with which the system can readily extended for a variety of applications requiring different spatial resolution.