Engineering Biosensors and Biomedical Detection Devices from 3D-Printed Technology

Limitation of 3D construction ability, complex preparation processes and developing customer demands have promoted people to find low-cost, rapid prototyping, and simple operation methods to produce novel functional devices in the near future. Among various techniques, 3D-printed technology is a promising candidate for the fabrication of biosensors and biomedical detection devices with a wide variety of potential applications. This review offers four important 3D printing techniques for biosensors and biomedical detection devices and their applications. The principle and printing process of 3D-printed technologies will be generalized, and the printing performance of many 3D printers will be compared. Despite the resolution restrictions of 3D-printed, these technologies have already shown promising applications in many biosensors and biomedical detection devices, such as 3D-printed microfluidic devices, 3D-printed optical devices, 3D-printed electrochemical devices, and 3D-printed integrated devices. Some of the most representative examples will also be discussed here, demonstrating that 3D-printed technology can rationally design biosensors and biomedical detection devices and achieve important applications in microfluidic, optical, electrochemical, and integrated devices.

Providing effective diagnostic information is the main function of biosensors and biomedical detection. 21,22 The importance of biosensors and biomedical detection is that they are closely related to human life and health. [23][24][25] An illustration of its potential is the detection of COVID-19, 26,27 which clearly tells whether people are infected. Tumor marker detection, 28,29 gene detection, 30,31 drug detection 32,33 and physiological information monitoring 34 have attracted much attention. Although many detection devices have been used clinically and achieved attractive results currently, some disadvantages of the preparation technology need to be improved. Traditional manufacturing processes, such as casting, stamping, photolithography, and milling requiring professionals to operate in a specific environment, time-consuming, and limitation of 3D construction ability, restrict the development of biosensing and biomedical detection devices. 34,35 3D-printed technology is ubiquitous in the field of biosensors and biomedical detection. Each 3D printing technique has different characteristics and application scenarios. 36 For example, i3DP has good resolution and biocompatibility, and can be used to prepare microfluidic devices. 37,38 FDM is fast and low-cost, and often used to prepare device supports and housings. [39][40][41] SLA has high resolution, good surface finish, and is used to making of master. 42,43 DLP has fast forming and high resolution, and is used to print microstructures. 44,45 Importantly, these 3D-printed methods have the advantages of one step from model to product, 3D construction, and complex structure shaping capability. 46 These capabilities play a critical role in transitioning from the laboratory to a commercial product. 47 3D-printed technology will become the main manufacturing method for future biosensors and biomedical detection devices. 48 This review summarizes the principles, printing process, and printing performance of 3D-printed technologies. 3D-printed technology has promising potential for biosensors and biomedical detection. Some exciting examples for the application of 3D-printed biosensors and biomedical detection devices such as microfluidic devices, optical devices, electrochemical devices, and integrated devices (Scheme 1) will be presented. With the rise of personalized devices and precision medicine, 3D-printed biosensors and biomedical detection devices will be a valuable addition.

Selection of 3D Printing Techniques for Biosensors and Biomedical Detection Devices
3D printing, also known as additive manufacturing, is a technology based on digital model files and using adhesive materials to construct objects through layer-by-layer printing. 43 The 3D printing process might have the opportunity to become a powerful technique for the fabrication of biosensors and biomedical detection devices. According to different 3D printing principles, four important 3D printing techniques, including i3DP, FDM, SLA, and DLP are summarized (Fig. 1). 35 No matter the printing technique, the printing process can be broken down into four steps: design, slicing, printing, and postprocessing (Fig. 2). It is significant to choose a suitable 3D printing technique to prepare biosensors and biomedical detection devices. Therefore, this section mainly introduces the principle, printing process, and printing performance of 3D printers. Table I summarizes the model, manufacturer, printing layer thickness, printing size, and application of 3D printers. In addition, the advantages and disadvantages of four 3D printing techniques are introduced.
3D printing principle and printing process.-The development of i3DP can be traced back to the invention of the inkjet 3D printer in 1976. The principle of i3DP is to generate ink droplets through z E-mail: bohu@xidian.edu.cn; geguanqun@126.com; shifangy@fmmu.edu.cn thermal or piezoelectric elements, and the ink droplets are sprayed onto the printing platform. 49 Solidifying layer by layer under ultraviolet radiation. After one layer is printed, the platform drops and starts a new layer. At present, multi-jet modeling (MJM) is the most important kind of i3DP, which can print two or multiple materials at the same time. The printing process based on MJM is divided into three steps. First, building the model in 3D design software and exporting the STL file format, then setting the printer parameters, and finally starting 3D printing. 50 During the printing process, two different print heads are used, one for photosensitive build material, one for sacrificing the support material, and finally remove the support material. Removal of inner and outer sacrificial materials in the microfluidics includes different methods, which have been discussed in detail by our group. 51 3D printers based on MJM mainly come from two companies, 3D Systems and Stratasys. The materials suitable for MJM include photocurable plastic resin, MED610 photopolymer, and VisiJet M3 Crystal. [52][53][54] FDM has become more and more mature and has important applications in many fields. In 1988, Scott Crump invented FDM.
The following year, Scott Crump established Stratasys, and FDM molding technology was patented by Stratasys. 55,56 In 1992, the first 3D printing product based on FDM technology was sold. The principle of FDM is to melt the filamentous hot melt material and push it out through a tiny nozzle. The nozzle can move along the X axis, and the worktable moves along the Y axis and Z axis. The melted hot melt material flows through the nozzle and solidifies in the set area. After one layer is deposited, the platform moves down, and the extrusion nozzle will deposit the next layer. 57 The FDM printing process is mainly divided into four steps. First, making a 3D data model, exporting the STL file format. Then using slicing software to process layers and automatically adding support. Next, setting the printing parameters and starting printing. Finally, device post-processing including detaching from support and surface polishing. [58][59][60] There are many companies that make 3D printers based on the FDM technology, such as Stratasys, Makerbot, Flashforge, Sethi3D, Creality, and Wanhao. Polylactic acid (PLA) and acrylonitrile butadiene styrene copolymers (ABS) are the most commonly used 3D printing materials using FDM technology. [61][62][63] SLA, as the earliest molding technology, was the first to be proposed and commercialized. Its development can be traced back to the concept of using a laser to find the surface of photosensitive resin and solidify to make three-dimensional objects put forward by Charles W. Hull in 1980s. 64,65 The 3D printing principle based on SLA technology is the liquid photosensitive resin, which is filled into the container. The laser beam is used to scan the liquid photosensitive resin surface line by line based on the layered cross-section data of the workpiece. Next the photosensitive resin is cured to form a thin layer. After the curing is complete, the worktable will move down and cover the original surface with a new layer of liquid resin. 66,67 The SLA printing process is mainly divided into three steps. The first step is to design, establish 3D model drawings, export STL format files, set printing parameters, and send instructions to the printer. Second, after confirming that the settings are correct, 3D printing will be started, and the material will be automatically filled into the ink cartridge. Finally, the parts need to be washed in isopropanol to remove the uncured resin from the Scheme 1. Schematic illustration of the 3D-printed devices as a potential application of the biosensors and biomedical detection.
surface. After washing, the parts need to be cured and polished again to make the surface smooth. 68,69 The main manufacturers are FORMLABS, 3D Systems and 3D Factory. Resin materials, WaterShed XC11122 and clear photopolymer, are used in the SLA technique. 48 DLP is recognized as the second generation of UV curing technology, which has a history of more than 20 years. 96 The principle of 3D printing based on DLP technology is similar to that of SLA 3D printing. In the curing process, the pattern of this layer is projected by a digital micromirror device (DMD) instead of a moving laser. The designed pattern is projected by DLP, which avoids the influence of spot size on printing performance. 79,97 The DLP printing process is mainly divided into three steps: first, designing the model and editing the slice of the model; second, downloading the model and starting printing; and finally, washing the printed device with alcohol and drying it with paper. The main    3D printing performance.-The key elements that affect the performance of i3DP are printing material, substrate properties, printing platform, and droplet generation. 99 The viscosity of the ink material is low enough to ensure the spray stability of the droplets. The direction and uniformity of the ink droplets and the roughness of the substrate surface affect the print resolution. The alignment and calibration of the platform are also affecting the print resolution of the device. 100 The i3DP has the advantage of being able to make two or more material combinations simultaneously, therefore it can print colorful devices. 101 In addition, this 3D printing technique has high accuracy and smooth surface. The disadvantages of this 3D printing technique are that it is easily affected by temperature and the removal of sacrificial materials. The i3DP has been widely applications in the field of biosensors and biomedical detection. The resolution and print size depend on the model of the printer. For example, the ProJet MJP 3600HD printer from 3D Systems has the resolution up to a layer thickness of 16 μm and a print size (x/y/z) of 298 × 185 × 203 mm. 102 The key elements that affect the performance of FDM are building temperature, slice height, nozzle diameter, air gap, raster width, and raster angle. 103 The building temperature affects the material deposition, the slice height and nozzle diameter affect the printing resolution, and the air gap, raster width, and angle affect the forming size. 104 The advantages of FDM technology are simple operation, low maintenance cost, and safe operation. There are many kinds of materials, such as engineering plastics ABS, polycarbonate (PC), polyphenylsulfone (PPSF), medical ABS, and their colored ones. Raw materials are usually supplied in the form of rolls for easy handling and quick replacement. The disadvantage is that the precision is low, the surface is not smooth and the forming speed is slow. 105 Printers based on FDM technology have a low resolution, about a few hundred microns. The print size is generally less than 400 × 400 × 400 mm. 106 The key elements that affect the performance of SLA are the laser, resin material, printing procedure, post curing spot diameter and placement direction. 103 The laser power has a great influence on the shape stability, and the shrinkage decreases with the increase in the laser power. Printing procedures such as printing speed and layer thickness will affect dimensional accuracy. Post-curing affects the surface roughness of the device. While the spot diameter has an impact on printing accuracy, a small spot has a negative impact on printing effectiveness and success rates. Because movement speed and displacement vary depending on the direction, the placement direction will also have an impact on printing accuracy. 107 3D printing technology based on SLA technology has many advantages, such as easily building large parts, surface finish, high accuracy and reused of uncured materials. The disadvantages are slow construction process, easy to be affected by temperature and uncertain mechanical properties. The resolution of printers based on SLA technology depends on the size of the laser spot, about 50 μm or more. The printing size is also different. The SLA750 printer from 3D Systems can print the size up to 750 × 750 × 550 mm. 108 Table II. The test of four 3D-printed methods to print different models. Four different models, including rectangular sections, straight channels, curved channels and serpentine channels. The rectangular sections, straight channels and curved channels all have six sizes. The height and width of the channels are 200 μm × 400 μm. Re-printed with the permission from Ref. 35. Copyright © 2020, Royal Society of Chemistry.    The key elements that affect the performance of DLP are light source, DLP, printing platform and post curing. 103 The intensity of the light source and DMD affect the printing accuracy. The printing size is limited by DMD. The operation of the printing platform and post-processing will affect the surface roughness. Compared with the SLA technique, the DLP technique has a faster printing speed and better 3D printing resolution. It also has the advantages of smooth surface and many printing materials. However, the disadvantages of 3D printing with the DLP technique are insecurity of the material and difficult to print large structure. The resolution of 3D printing of the DLP technique can reach up to 25 μm. Due to the limitations of DMD, the printing size is small, and the Fig. 4 production from 3D systems has a printing size of 124.8 × 70.2 × 346 mm. 109 The key elements affecting the printing performance of selective laser sintering (SLS) are laser power, scanning speed, scanning spacing and powder thickness. The density of printing devices rises as a result of the promotion of atom mobility across materials caused by an increase in laser power. The powder material absorbs less energy at a faster rate when the laser is scanning, which makes it harder for the sintering process to finish and lowers the density of the sintered pieces. The scanning spacing reduces the bonding degree of the two adjacent lines, which affects the strength after forming. In addition, the excessive thickness of the powder will also affect the strength after molding, which is due to the incomplete sintering of the bottom material. SLS is based on the principle that powdered materials are sintered under laser irradiation. The powder is spread evenly on the piston at the beginning of the print. The computer controls the scanning trajectory of the beam according to the slice model. Selective sintering of solid materials to form a layer of the part. After finishing a layer, the piston reduces the layer thickness by one, repeats the process above, and then obtains the three-dimensional component.
Our group have designed and printed standard 3D printing models for the testing of four 3D-printed techniques, including i3DP, SLA, DLP and FDM. Four different channels have been used, which include rectangular sections, straight channels, curved channels and serpentine channels. 35 The rectangular sections, straight channels and curved channels all have six sizes (Table II). The results show that i3DP has the best printing performance, with a cross section of 0.1 × 0.2 mm and a channel of 0.2 × 0.2 mm.
3D-printed materials.-Since the emergence of 3D printing technology, it has promoted the development of biosensors and biomedical detection devices. From a basic point of view, for each particular application, the materials available for the 3D printing process must be considered. 3D printing materials are versatile, which comes from the diversity of 3D printing systems. Here, the materials commonly used in four printing methods are summarized. The advantages and disadvantages of printing materials are generally discussed, as are their uses in biosensors and biological detection (Table III).
3D-printed efficiency.-Printing efficiency is one of the elements that must be taken into account while constructing biosensors and biological detection equipment. Table IV lists the factors that affect printing efficiency. Inkjet 3D printers use a series of inkjet heads to deposit tiny droplets that build and support materials, forming objects layer by layer. The print head has hole arrays that can spray droplets of both the construction material and the supporting material to build a 3D structure. The most common types of materials used in FDM 3D printing are thermoplastics. In the sprinkler, the substance is heated and melted. The sprinkler moves from point to line to face along the part profile and fill trajectory. After extrusion, the melted material quickly hardens and is attached to the adjacent material. SLA 3D printing is a dot projection when printing, which is limited to a single spot and needs to be printed gradually from point to line, from line to surface. DLP printing speed, in general, is one of the benefits of DLP 3D printing, which is especially clear when compared to others 3D printing technology. DLP 3D printing can project and aggregate an entire layer when printing. Generally speaking, theoretically, DLP printing speed is the fastest, inkjet printing is the second, and SLA is the slowest. The printing rate, however, is frequently correlated with the printing precision, printing algorithm, printing materials, and complexity, and the printing efficiency of various printer models will vary greatly.

3D-Printed Biosensors and Biomedical Detection Devices
3D-printed devices are easily adapted for the biosensors and biomedical detection due to its powerful performance, such as 3D shaping, one-step molding, and complex structures. 3D-printed microfluidic devices, 99 optical devices, 115 electrochemical devices [116][117][118] and integrated devices 119,120 prepared by different 3D printing techniques have become research hotspots. 3D printing technology has broadened the development of the field of microfluidics. 3D-printed microfluidic devices such as inertial focusing and enrichment tumor cell chips, nucleic acid detection chips, 3D cell culture devices, micro droplet generation devices, micro mixers, micro pumps, micro valves, etc. 74,121-128 have been successfully developed and applied. The SERS detection platform, fluorescent sensing platform, portable microscope, bioluminescent sensor and other 3D-printed optical devices 129,130 combined with optical equipment improve the detection performance. 3D-printed electrochemical biosensors for biochemical molecular detection, wearable physiological information monitoring, and in situ monitoring of cells are highly sensitive, low-cost, and portable and are crucial in the construction of micro-monitoring devices. [105][106][107]131,132 A detection platform often requires the combination of multiple functional modules. 3D-printed integrated devices are suitable for POCT, 16,27,31 such as PCR systems including the modules of pressure sample pretreatment, nucleic acid extraction, drive, heating, and detection. 3D-printed microfluidic devices.-The operational complexity, high expense, difficulty preparing complex structures, and stringent environmental requirements of conventional microfluidic devices are drawbacks. These flaws are made up for by the development of 3D printing technology, which is good news for scientists. Inspired by the bed topography in river meanders (Figs. 3a and 3b), our group have designed a novel 3D-printed river meander-inspired microfluidic chip for inertial focusing and enrichment. 72 The channel of the microfluidic chip is helical. Figure 3d shows the design model of helical channel, and Fig. 3e shows 3D-printed river meander-inspired microfluidic chip. A helical 3D micro-channel was printed using a commercial 3D printer based on the i3D technique, which has the characteristics of high resolution. VisiJet M3 Crystal material is highly biocompatible matrices, which is suitable for biosensor and biomedical detection. A numerical simulation microfluidic chip model is established to calculate the velocity field and vortex field in COMSOL (Fig. 3c). The experimental and simulation results are highly unified. Compared with other regular shapes, the river meander-like cross-section focus in a shorter time (Fig. 4a). Finally, a microfluidic chip with three outlets was designed. MDB-MB-231 cells were added to diluted rabbit whole blood to verify the performance of the chip. Figure 4b shows that MDB-MB-231 cells are collected by the intermediate outlet. The recovery rate of 85.4% (Fig. 4c) and the enrichment ratio of 1.86 (Fig. 4d) in whole blood can be achieved. This paper reports a unique river meander-like cross-section by using the ability of 3D printing high resolution and constructing complex structure. The 3D printing microfluidic device is successfully applied to the focusing and enrichment of inertial fluid.
3D-printed microfluidic devices have been developed for the nanomaterial printing. Although nanomaterial printing plays an important role in the fields of biosensors and biomedical detections, specifically printing methods and materials limit the applications. Our group have reported a capillary force-driven stamped approach (CFDS) based on 3D-printed microfluidics devices for nanomaterial printing (Figs. 5a, 5b and 5c). 73 The main principle is loading and deposition of the printing material by using the capillary force provided by the microchannel of the microfluidic device and the filter paper fibers. The CFDS approach has three steps (Fig. 5d). Firstly, 3D-printed microfluidic device, produced by the i3D technique, is used to fill the microchannel with nanomaterials aqueous solution through the action of capillary force (Fig. 5e). Then, the stamper is placed on paper substrate to deposit nanomaterials (Fig. 5f). Finally, the stamper is removed and the paper substrate with nanomaterials (Fig. 5g) is customized into the desired shape. The advantage of 3D-printed microfluidic device is that it can print many patterns according to its own needs. The printing performance by designing various angles and patterns has been verified (Fig. 6a). More importantly, it can be used to print SERS detection substrates. The SERS signal of 4-MBA was measured on the patterned array prepared by the CFDS approach (Fig. 6b), which have shown the great potential for SERS detection (Figs. 6c-6d). In a word, a 3D printing microfluidic stamper and paper substrate are combined to create a device for directly printing patterned nanomaterial aqueous solutions.
3D-printed technology has been widely used in the microfluidic field, but 3D-printed microfluidic chip based on the droplet polymerase chain reaction (PCR) is rarely reported. Our group have reported a 3D-printed microfluidic chip for spatial domain PCR systems by the i3DP technique (Fig. 7a). 133 The chip consists of a T-junction for producing oil-in-water droplets (Fig. 7b) and a serpentine channel for the PCR process. The droplets flow through high and low temperature regions to achieve temperature cycling. Figure 7c shows the infrared temperature image on the chip. A 400 mm wide and a 500 mm length serpentine microchannel were   (Figs. 7d and 7e). For droplet-based miRNA-21PCR detection, a microfluidic chip is created using 3D printing. This demonstrates the enormous potential of 3D printing technology for microfluidic research.
3D-printed optics devices.-The precise positioning of optical and detecting equipment is necessary due to the complexity of the optical path construction. The greatest option is without a doubt 3D printing, which can not only produce intricate optical element support structures but also fix the detection item in a particular location. Drug detection in the biofluids is an indispensable part of clinical diagnosis. At present, most detection methods require large instruments and complex operation. Our group have designed and produced a low-cost portable 3D-printed paper cartridge for quantitative detection of drugs in biofluids (Fig. 8a). 79 3D-printed cartridge was fabricated by the high precision FDM 3D printer. The 3Dprinted paper cartridge mainly consisted of paper tip, 3D-printed cartridge, sampler and cover, which cost only one dollar (Fig. 8b). The detection step is simple, roughly divided into three steps (Fig. 8c). The first step is to add the sample and drop the sample solution into the groove of the sampler cover. The second step is that the sample is pre-concentrated at the tip of the paper, followed by SERS detection under a portable Raman spectrometer. The sample was pre-concentrated through the 3D-printed paper cartridge. SERS performance achieved 9.93-fold improvement and the whole detection process took only 1 h. The anticancer drugs of epirubicin hydrochloride, cyclophosphamide and their mixtures were quantitatively detected successfully in the bovine serum. Figures 9a and 9b show the limit of detection (LOD) of epirubicin hydrochloride in serum is × − 5 10 8 M. Selective concentrations of epirubicin hydrochloride in serum from × − 8 10 7 to × − 1 10 5 M for precise quantification (Figs. 9c and 9d). The precise positioning of optical and detecting equipment is necessary due to the complexity of the optical path construction. The greatest option is without a doubt 3D printing, which can not only produce intricate optical element support structures but also fix the detection item in a particular location.
SERS detection of cells is of great significance in the biomedical research. At present, there is lack of a low-cost and rapid detection method. Our group have built a simple SERS dynamic fluid detection platform, which mainly includes soft tubular, 3D-printed template and connectors. 134 3D-printed template was easy to obtain and cheap, and it can be used to build a soft tubular microfluidic detection platform (Figs. 10a and 10b). The dynamic liquid detection platform has the advantages of dynamic mixing and continuous spectrum acquisition, and the detection results have good stability and repeatability. Three different cell lines are investigated on this platform and classified by the model based on K-nearest neighbor (K-NN) algorithm (Fig. 10c). The results have shown that the sensitivity was more than 83.3%, the specificity was more than 91.6%, and the accuracy was 94.1% ± 1.14%. This method has the potential applications in clinical diagnosis and cell research. In addition, our group have further developed this method to produce a 3D hydrodynamic focusing Raman platform (Fig. 11a). 78 The platform assembled by a coaxial needle, a quartz capillary, a 3D printing holder and soft silicone tube (Fig. 11b). This fluid focusing detection method can be used for the detection of single particle without complex operation. Using Poisson distribution analysis and dynamic single particle Raman detection, the classification model of KNN algorithm is established, and the classification accuracy, sensitivity and specificity were 100%. Figure 11c shows the classification results of three experiments, in which colors represent different λ 0 and each peak is the SERS spectrum of particles at − 1001 cm . 1 The advantages of 3D-printed holder to build a testing platform were low cost and the shape can be designed according to the requirements.
Limited medical resources in impoverished areas severely limit the ability to diagnose diseases and lead to high mortality from treatable diseases. Fluorescence microscopes are expensive and not portable, so it is imminent to develop a low-cost POCT fluorescence detection platform. Stephanie Knowlton et al. have proposed a fluorescence imaging and magnetic focusing platform for cell sorting (Fig. 12a). 113 The platform utilizes cameras of smartphone for cell imaging in brightfield, darkfield, and fluorescence imaging modes. Fluorescence imaging mode can eliminate a population from the image. Schematic of optical path and microscope under brightfield conditions have been shown in Fig. 12b. Figure 12c shows a schematic diagram of the light path and a top view of a microscope in a darkfield imaging configuration of a sample excited by blue light and emitting green light. The A549 cancer cells and 3T3 mouse fibroblasts were successfully stained with acridine orange and imaged with a smartphone-based fluorescence detection platform. Because rapidly dividing cells emit more red light, when excited with blue light, cells in the red channel have a higher signal than the green channel.
Single-molecule analysis has been considered as the next frontier in bioanalysis and diagnostics. James W. P. Brown et al. have reported a portable 3D-printed microscope for single-molecule detection. 135 3D printing-based plug-and-play single-molecule confocal system has been used for the detection of Parkinson's disease biomarkers. The system was performed by the photon counting histograms (PCH) . Figures 13a-13c show a simple singlemolecule confocal system with all optical components embedded in a 3D-printed housing. Single molecules are detected as they diffusing into and out of the focal volume (Fig. 13d). Compared with water alone, Alexa-568 labelled α-syn has very high signals, and the authors experimentally demonstrated that a 3D printingbased portable microscope can detect individual protein molecules at concentrations below 100 pM (Fig. 13e). In addition, the system can determine the degree of oligomerization of molecules, and Fig. 13f shows that green fluorescent proteins (GFP) of different structures can be distinguished. The three curves shown in Fig. 13g are free Alexa-488, α-synuclein labeled with Alexa-488, and α-syn labeled with Alexa-488 in excess sodium dodecyl sulphate (SDS).
3D-printed electrochemical devices.-3D-printed technology has been widely used in the field of electrochemical sensors. Kanyapat Teekayupak et al. have reported a 3D-printed electrochemical sensing device for non-enzymatic detection of creatinine. 106 The preparation scheme of the 3D-printed electrodes (3DE) is shown in Fig. 14a. FDM 3D printer was used to print the electrodes, then the reference electrodes were coated with Ag/AgCl ink. All the printed electrodes were glued to the PVC substrate, and then the electrodes were coated with insulators and hydrophobic treatments. Finally, the electrodes were activated with 0.5 M sodium hydroxide (NaOH) (Fig. 14b). The authors combined a smartphone with a 3D-printed electrochemical sensor to detect the level of creatinine in urine. The detection performance of a portable smartphone was Verified. The current increased with the concentration of creatinine, which was in the range of 0.5-30.0 mM. Two linear ranges were observed in the range of 0.5-5.0 mM and 7.0-30.0 mM. The author guessed that it may be the van der Waals interaction between the 3DE surface conductive material and the creatinine heterocyclic structure.
3D-printed electrochemical devices can monitor cells and molecules in situ without sample extraction. Pradeep Ramiah Rajasekaran et al. proposed a system with 3D-printed modular and integrated electrochemical sensors, which can detect physical and biochemical information non-invasively in real time. 111 The devices consist of 3Dprinted bottom chamber, electrode-integrated PETE membrane, access port, top chamber and electrodes connecting to the sensor contact pads. The most important part of the device is the multimodal    can endow real-time in situ sensing capabilities in the cell culture system. As the cell culture time increases, the impedance decreases exponentially. Detection of ferrocene dimethanol released from the basolateral side using CV detection method, in which molecules diffuse through electrode-integrated porous membranes mimic biomarkers.
The demand for wearable sensing devices with non-invasive detection continues to increase, and Vassiliki Katseli et al. reported a 3D-printed wearable glucose monitoring device. 110 The device uses an electrochemical ring (e-ring) made by a dual-extruder 3D printer combined with a smartphone to monitor the glucose content in human sweat (Fig. 15a). Thermoplastic polyurethane (TPU) has excellent flexibility and durability, which has been widely used in the medical industry (Figs. 15b and 15c). The e-ring is composed of a TPU holder and three electrodes. Figure 15c verifies the long-term stability, three measurements per day for 10 days, and the 3D-printed electronic ring remains stable. The glucose in the sweat was monitored within one hour and two hours after the meal. The glucose content first increased and then decreased within two hours after the meal.
3D-printed integrated devices.-3D-printed integrated devices have been widely applied in biomedical detection and biosensor. However, there are few reports on multi-function devices with POCT. Chao Liang et al. describe a film-lever actuated switch technology combining with 3D-printed device, elastic film and peripheral control module. 136 3D-printed integrated devices combine multi-functional dispensing, on-demand releasing, long-term reagent storage and robust operation to enable rapid testing without the professionals and strict environments (Figs. 16a-16c). The working principle of the platform is mainly to apply pressure to break the sealed chamber formed by the lever, and the elastic film to drive the fluid out. Lever lengths are dependent on the critical pressures by the manipulation of liquids. The teams then developed a polymerase chain reaction (PCR) multiple detection system, including sample pretreatment, reagent mixing, reaction and detection. The performance of sample-in-answer-out was realized (Figs. 16d-16f). The systems detect influenza virus A (IVA) and influenza virus B (IVB) separately. The test results are shown in the 3D-printed integrated PCR system has a LOD of 100 copies ml −1 . Compared with the commercial PCR system, it has better detection performance. 3D-  printed technology has the potential to provide help for POCT systems.
Rapid and accurate detection of the RNA of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is very important during the spread of the virus around the world. SLA is the most popular printing technology, with high printing accuracy and a wide range of materials to choose from. Photosensitive resin materials with different properties can be selected according to different application scenarios, such as hard resin, high temperature resistant resin, low viscosity resin, and biocompatible resin. Using SLA technology, Devora Najjar et al. created a 3D printing chip lab that combined electrochemical modules to fully detect RNA in serum and saliva. 91 A 3D-printed lab on a chip, including reagent storage chamber, sample preprocessing chamber, reaction chamber and electrochemistry sensor chip (Figs. 17a and 17b). Then, the author improved the electrochemical sensor, using four electrodes to detect three antigens and RNA. The analytical performance of the electrochemical sensor was verified by testing the combination of four clinical samples, including serology, RNA negative and positive. The chip can simultaneously detect SARS-CoV-2 RNA in saliva and anti-SARS-CoV-2 immunoglobulin in plasma within 2 h. The device automatically extracts, condenses, amplifies and detects viral RNA from untreated saliva samples. The whole process has no manual operation, high integration and low cost. It has a wide application prospect in the field of POCT.
In order to avoid the potential adverse effects of pesticides on people, Xiaofan Ruan et al. reported a dual-functional optical platform (DFOP) based on 3D-printed lateral flow immunoassay (LFIA) to detect the main metabolite and biomarker of atrazine exposure diaminochlorotriazine (DACT). 137 The biosensors utilize Au@PtPd nanoparticles as a signal indicator to provide a rapid chromatographic readout inversely proportional to the analyte concentration (Figs. 18a-18c). 3D-printed DFOP were prepared by FDM printer. Intelligent optical detection is realized by combining  smartphones and image analysis software with biosensor (Fig. 18d). In order to verify the detection performance of 3D-printed DFOP, the detection signal of DACT is quantized. There is a good linear relationship between the color intensity and the concentration of DACT. In addition, the strips are placed in the substrate system to terminate the reaction. 3D-printed integrated devices are more and more popular among researchers in the field of biomedical detection and biosensors because of their superior performance and easy access.

Conclusions and Perspectives
In summary, biosensors and biomedical detection devices fabricated by 3D-printed technology have been discussed. The different 3D printing principles and materials result in different kinds of printing performance. The iD3P printing methods are suitable for the preparation of microchannels. The DLP and SLA printing methods are suitable for the preparation of microstructure, and the FDM printing method is suitable for the preparation of holder and housing. In terms of resolution, i3DP is the highest and FDM is the lowest. From the point of view of printing costs, FDM is the lowest. Many biosensors and biomedical detection devices use 3D-printed technology, such as 3D-printed microfluidic devices, optical devices, electrochemical devices and integrated devices. 3D-printed microfluidic devices have been used for cell focusing and enrichment, nanomaterial printing and nucleic acid detection. 3D-printed optical devices have been used to build portable microscopes, SERS and fluorescence detection platforms. 3D-printed electrochemical devices have been used for biomolecule detection, cell and human sweat information monitoring. 3D-printed integrated devices are used for nucleic acid and antibody detection of viruses. It is believed that 3D-printed technology will play an increasingly important role in the fabrication of biosensors and biomedical detection devices in the future.
Although the 3D-printed technology has achieved great success in the fields of biosensors and biomedical detection, there still remain some challenges and questions. How to reduce printing costs? Can we find a simpler way to remove the support material? How to improve the printing resolution? How to increase the print size? How to achieve mass production? How to increase the strength of the print? How to enrich printing materials? Overcoming these challenges and difficulties in the future will further promote and strengthen the capability of 3D-printed for biosensor and biomedical detection devices.